UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Impaired attentional processing as a risk factor for falls in older adults Nagamatsu, Lindsay 2013

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2013_fall_nagamatsu_lindsay.pdf [ 2.06MB ]
Metadata
JSON: 24-1.0165517.json
JSON-LD: 24-1.0165517-ld.json
RDF/XML (Pretty): 24-1.0165517-rdf.xml
RDF/JSON: 24-1.0165517-rdf.json
Turtle: 24-1.0165517-turtle.txt
N-Triples: 24-1.0165517-rdf-ntriples.txt
Original Record: 24-1.0165517-source.json
Full Text
24-1.0165517-fulltext.txt
Citation
24-1.0165517.ris

Full Text

  IMPAIRED ATTENTIONAL PROCESSING AS A RISK FACTOR FOR  FALLS IN OLDER ADULTS   by  LINDSAY NAGAMATSU  B.A., The University of British Columbia, 2006 M.A., The University of British Columbia, 2009     A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY   in   THE FACULTY OF GRADUATE STUDIES  (Psychology)    THE UNIVERSITY OF BRITISH COLUMBIA  (Vancouver)   September 2013   ? Lindsay Nagamatsu, 2013     ii Abstract Falls are a leading health care issue in our rapidly aging society. While impaired cognitive functioning has previously been linked to falls, my dissertation aims to understand the specific relationship between attentional processing and falls/falls risk. My dissertation is comprised of four separate studies, each examining a specific facet of attention and its corresponding association with falls. The specific domains of attention that I examine include: 1. Visual-spatial attention in the context of task-irrelevant peripheral stimuli; 2. Dual-task ability under cognitive load using a virtual reality system intended to simulate real-world experiences; 3. Time spent engaged in task-unrelated thoughts (i.e., mind-wandering); and 4. The neural correlates that subserve selective attention and executive control. Using multi-modal measures, including event-related potentials, functional magnetic resonance imaging, and behavioural performance, my research examines how each area of attention is impaired as a function of falls history and/or physiological falls risk. Based on the attentional resource model, my research converges on the notion that falls risk is associated with a reduction in the level of available attentional resources, in addition to an impaired ability to appropriately allocate resources to the primary task. Importantly, such deficits in attentional processing may contribute to postural instability, given that the control of balance and posture require a greater proportion of cognitive resources with age. Taken together, these findings have critical implications for developing novel intervention strategies aimed towards improving quality of life and independence among older adults.    iii Preface I prepared the content of this dissertation, with minor edits from Todd Handy and Teresa Liu-Ambrose. The research presented in Chapters 2 to 5 was primarily conducted by myself and all studies have been published or submitted for publication (details below).  A version of Chapter 2 has been submitted for publication as Nagamatsu, LS, Munkacsy, M, Liu-Ambrose, T, and Handy, TC. (submitted). Altered attention to task-irrelevant stimuli is associated with falls risk in seniors. I was responsible for study conception and design, data analysis and interpretation, and manuscript composition. M Munkacsy was primarily responsible for participant recruitment and data collection. T Liu-Ambrose and T Handy were responsible for study conception and design, interpretation, and critical review of the manuscript. This study was approved by the University of British Columbia?s Research Ethics Board, C04-0161 (H04-70161-002): Cortical Networks of Attentional Orienting.  A version of Chapter 3 has been published as Nagamatsu, LS, Voss, M, Neider, MB, Gaspar, JG, Handy, TC, Kramer, AF, and Liu-Ambrose, TYL. (2011). Increased cognitive load leads to impaired mobility decisions in seniors at risk for falls: A virtual reality study. Psychology and Aging, 26, 253-259. I was responsible for study conception and design, participant recruitment, data collection, analysis, and interpretation, and manuscript composition. M Voss, MB Neider, JG Gaspar, TC Handy, AF Kramer, and TYL Liu-Ambrose were responsible for study conception and design, interpretation, and critical review of the manuscript. This study was approved by the University of Illinois and Champaign-Urbana?s Institutional Review Board, 08106: Understanding Attention in Realistic Tasks and the University of British Columbia?s Research Ethics Board, H13-01864: CAVE study.   iv A version of Chapter 4 has been published as Nagamatsu, LS, Kam, JWY, Liu-Ambrose, T, Chan, A, and Handy, TC. (in press). Mind-wandering and falls risk in older adults. Psychology and Aging. I was responsible for study conception and design, data analysis and interpretation, and manuscript composition. JWY Kam, T Liu-Ambrose, and TC Handy were responsible for study conception and design, interpretation, and critical review of the manuscript. A Chan was responsible for participant recruitment and data collection. This study was approved by the University of British Columbia?s Research Ethics Board, C04-0161 (H04-70161-002): Cortical Networks of Attentional Orienting.  A version of Chapter 5 has been published as Nagamatsu, LS, Hsu, CL, Handy, TC, and Liu-Ambrose, TYL. (2011). Functional neural correlates of reduced physiological falls risk. Behavioral and Brain Functions, 7(37), 1-9. I contributed to study conception and design and was responsible for data collection, analysis, and interpretation and manuscript composition. CL Hsu was responsible for data collection and interpretation. TC Handy was responsible for critical review of the manuscript. TYL Liu-Ambrose was primarily responsible for study conception and design, interpretation, and critical review of the manuscript. This study was approved by the University of British Columbia?s Research Ethics Board, H06-03216-0: Brain Power: Resistance Training and Cognitive Function in Older Women.   v Table of contents Abstract ......................................................................................................................................................... ii?Preface .......................................................................................................................................................... iii?Table of contents .......................................................................................................................................... v?List of tables .............................................................................................................................................. viii?List of figures ............................................................................................................................................... ix?List of abbreviations .................................................................................................................................... x?Acknowledgments ....................................................................................................................................... xi?Dedication ................................................................................................................................................... xii?CHAPTER 1: Introduction ............................................................................................................................ 1?Falls in Older Adults ................................................................................................................................................ 1?Risk Factors for Falls .............................................................................................................................................. 2?Falls and Cognition ................................................................................................................................................. 3?Basic cognitive processes ..................................................................................................................................... 3?Executive cognitive functioning ............................................................................................................................. 4?Dual-task performance .......................................................................................................................................... 6?Attention ................................................................................................................................................................... 9?Aspects of attention ............................................................................................................................................ 10?Overview of Dissertation ....................................................................................................................................... 14?Main Research Questions ..................................................................................................................................... 14?Methodology .......................................................................................................................................................... 14?Measures of falls and falls risk ............................................................................................................................ 15?Measures of attention ......................................................................................................................................... 15?Overview of Studies .............................................................................................................................................. 16?CHAPTER 2: Altered attention to task-irrelevant stimuli is associated with falls risk in seniors ...... 18?Introduction ............................................................................................................................................................ 18?Methods .................................................................................................................................................................. 21?Participants ......................................................................................................................................................... 21?Falls risk .............................................................................................................................................................. 22?Descriptive measures ......................................................................................................................................... 22?Procedure ........................................................................................................................................................... 23?Electrophysiological recording and analysis ....................................................................................................... 24?Data analysis ...................................................................................................................................................... 25?Results .................................................................................................................................................................... 26?Descriptive measures and falls risk .................................................................................................................... 26?Behaviour ............................................................................................................................................................ 28?Electrophysiology ................................................................................................................................................ 29?Discussion .............................................................................................................................................................. 36?Primary finding .................................................................................................................................................... 37?Secondary finding ............................................................................................................................................... 38?Additional issues ................................................................................................................................................. 39?Conclusion ............................................................................................................................................................. 40?CHAPTER 3: Increased cognitive load leads to impaired mobility decisions in seniors at risk for falls .............................................................................................................................................................. 42?Introduction ............................................................................................................................................................ 42?Methods .................................................................................................................................................................. 44?Participants ......................................................................................................................................................... 44?Procedure ........................................................................................................................................................... 44? vi Measures ............................................................................................................................................................ 45?Analysis ............................................................................................................................................................... 48?Results .................................................................................................................................................................... 48?Descriptive measures ......................................................................................................................................... 48?Computer dual-task performance ....................................................................................................................... 50?CAVE .................................................................................................................................................................. 51?Discussion .............................................................................................................................................................. 54?CHAPTER 4: Mind-wandering and falls risk in older adults: Evidence for failures in compensatory cognition ..................................................................................................................................................... 59?Introduction ............................................................................................................................................................ 59?Methods .................................................................................................................................................................. 64?Participants ......................................................................................................................................................... 64?Falls .................................................................................................................................................................... 65?Stimuli and procedure ......................................................................................................................................... 65?Electrophysiological recording and analysis ....................................................................................................... 67?Physical falls risk factors ..................................................................................................................................... 69?Results .................................................................................................................................................................... 69?Falls .................................................................................................................................................................... 69?Behaviour ............................................................................................................................................................ 70?Electrophysiology ................................................................................................................................................ 73?Physical falls risk factors ..................................................................................................................................... 76?Discussion .............................................................................................................................................................. 76?CHAPTER 5: Functional neural correlates of reduced physiological falls risk .................................... 81?Introduction ............................................................................................................................................................ 81?Methods .................................................................................................................................................................. 83?Participants ......................................................................................................................................................... 83?Randomization .................................................................................................................................................... 84?Exercise intervention ........................................................................................................................................... 84?Descriptive variables ........................................................................................................................................... 84?Dependent variable: physiological falls risk ........................................................................................................ 85?Independent variables of interest ........................................................................................................................ 85?Statistical analyses ............................................................................................................................................. 88?Results .................................................................................................................................................................... 89?Participants and variables of interest .................................................................................................................. 89?Linear regression model ..................................................................................................................................... 95?Discussion .............................................................................................................................................................. 96?CHAPTER 6: General discussion ............................................................................................................ 100?Summary of Studies ............................................................................................................................................ 100?Main Research Questions Revisited .................................................................................................................. 101?1. What specific types of attention are associated with falls/falls risk? ............................................................. 101?2. How might impaired attentional processing contribute to increased falls risk? ............................................. 103?3. What underlying neural structures are implicated in the relationship between attention and falls risk? ....... 105?Limitations ............................................................................................................................................................ 106?Future Directions ................................................................................................................................................. 108?1. What accounts for the visual-field asymmetries previously found in fallers? ................................................ 108?2. What underlying neural circuits, structures, and functions contribute to falls risk? ....................................... 109?3. Might impaired task-switching among fallers account for their observed attentional processing deficits? ... 109?4. Can we develop a simple and cost-effective test to evaluate executive cognitive function as a falls risk-factor? ............................................................................................................................................................... 110?5. How can we use the information we?ve gained to reduce falls risk? ............................................................. 111? vii Final Conclusions ................................................................................................................................................ 112?References ................................................................................................................................................ 114?    viii List of tables Table 2.1. Descriptive information. ............................................................................................................... 27?Table 2.2. Behavioural results. ..................................................................................................................... 28?Table 2.3. Mean amplitudes for P1 and N1 ERP components time-locked to task-irrelevant probes. ......... 31?Table 2.4. Mean amplitudes for P300 ERP component time-locked to central stimuli ................................. 35?Table 3.1. Descriptive measures. ................................................................................................................. 49?Table 3.2. Task performance in the CAVE and on the computer-based dual-task as a function of condition. ............................................................................................................................................................. 50?Table 4.1. Descriptive characteristics. .......................................................................................................... 70?Table 5.1. Descriptive statistics for variables of interest (N=73). .................................................................. 90?Table 5.2. Voxel Cluster Statistics from fMRI. .............................................................................................. 92?Table 5.3. Multiple linear regression model summary for improved physiological falls risk. ........................ 95?     ix List of figures Figure 2.1. Stimulus presentation and timing. .............................................................................................. 24?Figure 2.2. ERP waveforms for the P1 and N1 components. ....................................................................... 30?Figure 2.3 Correlations between mean ERP amplitudes (uV) and falls risk for the N1 for stimuli presented in the left visual field, measured at ipsilateral sites (OL+). ...................................................................... 33?Figure 2.4. ERP waveforms for the P300 component. ................................................................................. 34?Figure 2.5 Correlations between mean ERP amplitudes (uV) and falls risk for the P300 for targets presented at fixation during low load blocks. ....................................................................................... 36?Figure 3.1. A screen-shot of the virtual reality display. ................................................................................. 47?Figure 3.2.1. Street crossing performance as a function of condition (No distraction, Music, and Phone) and falls-risk group. .................................................................................................................................... 52?Figure 3.2.2. Street crossing performance as a function of condition (No distraction, Music, and Phone) and falls-risk group. .................................................................................................................................... 54?Figure 4.1. Trial sequence and timing of the SART paradigm. ..................................................................... 66?Figure 4.2. Scatterplots of significant correlations. ....................................................................................... 72?Figure 4.3. Grand averaged waveforms for our ERP data (n = 15). ............................................................. 74?Figure 4.4. Topographies for ERP components as a function of component and attentional state. ............. 75?Figure 5.1. The Flanker Task. ...................................................................................................................... 87?Figure 5.2. Brain regions demonstrating an increased hemodynamic response on incongruent relative to congruent trials. ................................................................................................................................... 93?    x List of abbreviations 6-MWT Six minute walk test ABC Activities-Specific Balance Confidence scale ACC Anterior cingulate cortex AD Alzheimer?s disease ADHD Attention deficit hyperactivity disorder BAT Balance and tone BMI Body mass index CAVE Cave virtual environment DMN Default mode network DTC Dual-task costs EEGs Electroencephalograms EOGs Electro-oculograms ERPs Event-related potentials FCI Functional comorbidity index fMRI Functional magnetic resonance imaging GDS Geriatric depression scale HAROLD Hemispheric asymmetry reduction in older adults MMSE Mini-mental status examination MoCA Montreal cognitive assessment OFC-In Orbital frontal cortex ? Insula  PASE Physical activities scale for the elderly PCG-ACC Paracingulate gyrus ? Anterior cingulate cortex PD Parkinson?s disease PPA Physiological profile assessment RCTs Randomized controlled trials RT Resistance training RTs Reaction times SART Sustained attention to response task SPPB Short physical performance battery TMS Transcranial magnetic stimulation TUG Timed up and go test VRE Virtual reality environment WWT Walking while talking     xi Acknowledgments My sincerest thank you to my mentors, family, husband, friends, collaborators, and colleagues who have supported my academic journey. My success is due to your guidance, patience, and encouragement.  My dedication to research during my PhD was possible thanks to trainee awards from the Michael Smith Foundation for Health Research (MSFHR) and the Natural Sciences and Engineering Research Council of Canada (NSERC).  Thank you to the research assistants involved in the research presented in this dissertation for their assistance with participant recruitment and data collection. I also thank my supervisors Todd Handy and Teresa Liu-Ambrose for their support with study conception and manuscript review.  Chapter 2: Funding for this study provided by a Discovery Grant from the Natural Sciences and Engineering Research Council of Canada (NSERC) to TCH, MOB-93373 from the Canadian Institutes of Health Research (CIHR) to TLA, and a CIHR Emerging Team Grant: Mobility in Aging to TLA and Karim Miran-Khan.  Chapter 3: This work was supported by The National Institute on Aging (R37 AG025667, RO1 AG25032) to AFK and the Canadian Institute of Health Research (MOB-93373) to TLA.   Chapter 4: Participants for this study were recruited from the Cognition and Mobility grant from the Canadian Institutes of Health Research (MOB-93373) to TLA.   Chapter 5: This work was supported by the Vancouver Foundation (BCM06-0035), the Michael Smith Foundation for Health Research Establishment Grant (CI-SCH-063(05-1)CLIN), and the Canadian Institutes of Health Research (MOB-93373) to TLA.        xii Dedication   ?Go confidently in the direction of your dreams. Live the life you have imagined.?  -Henry David Thoreau   ?Grow old along with me! The best is yet to be.?   -Robert Browning  1 CHAPTER 1: Introduction This dissertation aims to examine the relationship between impaired attentional processing and falls risk in older adults. Chapter 1 focuses on a review of the literature connecting cognition and falls, centered specifically on executive cognitive functioning and dual-task performance. I will then present a summary of the specific areas of attention my dissertation will cover, as well as briefly provide context for why these forms of attention are hypothesized to relate to falls risk. Next, I will turn to the main research questions that will be discussed in this dissertation, followed by an overview of my methodological approach for answering these questions.  Falls in Older Adults The number of seniors in our society is growing at an unprecedented rate, with those aged 60+ expected to triple worldwide to reach nearly two billion by 2050 (United Nations, 2002). One of the leading causes of loss of independence, disability, and death among our aging population is falls. Falls have serious health implications for older adults. In particular, 20% of falls in older adults require medical attention, with 5% of patients sustaining fractures (e.g., Tinetti, Speechley, & Ginter, 1988). At the societal level, falls result in over $2.4 billion annually in health care costs in Canada alone (The Hygeia Group, 1998). Furthermore, falls frequently result in reduced quality of life (e.g., Lord, Sherrington, Menz, & Close, 2007) and negative psychological outcomes, such as fear of falling (e.g., Tinetti, Deleon, Doucette, & Baker, 1994). With approximately one third of community-dwelling older adults experiencing one or more falls per year (e.g., Tinetti et al., 1988), understanding the key factors that contribute to increased falls risk is a growing heath-care priority.   2 Risk Factors for Falls In this dissertation, falls are defined as ?unintentionally coming to the ground or some lower level and other than as a consequence of sustaining a violent blow, loss of consciousness, sudden onset of paralysis as in stroke or an epileptic seizure? (Gibson, Andres, Issacs, Radebaugh, & Worm-Petersen, 1987). Although falls are often referred to as ?accidents?, evidence suggests that they are attributable to causal processes rather than as the result of truly random events (Grimley-Evans, 1990).  Demographically, falls are more likely to occur in women and in older adults living in residential care (e.g., Lord et al., 2007). Falls risk also increases with age and with a history of previous falls (e.g., Lord et al., 2007). Conventionally, falls are viewed as stemming from poor physical functioning, including impaired balance or reduced musculoskeletal strength (e.g., Lord et al., 2007). Other risk factors for falls, however, have been identified. For example, pharmacological risk factors include prescriptions for anti-psychotics or benzodiazepines (e.g., Lord et al., 2007). In addition, recent research has examined the contribution of impaired cognitive functioning as a risk factor for falls.   It has been established that cognitive impairment is a major risk factor for falls (e.g., Tinetti et al., 1988). Evidence suggests that even mild reductions in cognitive abilities increases physiological falls risk (e.g., Liu-Ambrose, Ashe, Graf, Beattie, & Khan, 2008). Furthermore, diagnosis of a neurodegenerative disease, such as dementia, Alzheimer?s disease (AD), or Parkinson?s disease (PD), is associated with a greater incidence of falls (e.g., Bloem, Valkenburg, Slabbekoorn, & van Dijk, 2001; van Doorn et al., 2003). Indeed, some studies have reported the occurrence of falls in patients to be up to three times higher (Morris, Rubin, Morris, & Mandel, 1987). In addition falls in patients may be more serious (i.e., injurious falls), more likely to reoccur, and may also lead to more negative outcomes, such as fear of falling, than in otherwise healthy older adults (e.g., Bloem, Valkenburg, Slabbekoorn, & van Dijk, 2001).  3 Falls and Cognition Early research on the relationship between falls and cognition focused on general tests of global cognitive function, such as the Mini-Mental Status Examination (MMSE). For example, in a prospective study, MMSE scores at baseline were found to predict rates of falling over an eight-year period (Anstey, von Sanden, & Luszcz, 2006). Although the MMSE is a widely used and easy-to-administer clinical tool, it assesses multiple cognitive domains, including attention, spatial ability, verbal skills, and memory, rendering it difficult to establish which specific areas of cognition are related to falls. In response to the MMSEs lack of specificity, numerous studies have been conducted to test individual components of cognition to determine the specific domains of cognition that are most related to falls. Such studies fall into three main categories and will be summarized below: 1. Basic cognitive processes; 2. Executive cognitive functions; and 3. Dual-task performance.  Basic cognitive processes In general, cognition can be defined as the mental functions that are required for information processing. More specifically, basic cognitive processes include perception, attention, memory, and learning. Research has found that falls are associated with impaired performance on tasks that tap into basic cognitive domains. For example, in the study by Anstey et al. (2006) mentioned above, the authors reported declines in verbal ability, processing speed, and immediate memory to be independently associated with increased falls.  The evidence supporting the link between reaction times (RTs) and falls is equivocal. First, non-significant between-group differences comparing fallers and non-fallers on both simple and choice RTs have been reported (e.g., Woolley, Czaja, & Drury, 1997). Second, Lord and Fitzpatrick (2001) found that senior fallers were impaired on a choice-stepping reaction time test, a task requiring participants to step on illuminated  4 floor tiles as quickly as possible. These contradictory results may be attributed to different task requirements. For example, choice reaction time tests are more difficult than simple reaction time tests. In addition, Lord and Fitzpatrick (2001) had participants use their feet for the reaction time test, which may be more directly relevant to falls. An alternative explanation is that reaction times per se may not be related to falls, but rather the consistency of reaction times. Specifically, Hausdorff et al. (2006) found that fallers display increased within-subjects variability in their reaction times. Indeed, high intra-individual variability is often associated with pathophysiology, and is attributed to deficits in functioning of the frontal lobes, which are essential for successful cognitive functioning (e.g., Bellgrove, Hester, & Garavan, 2004; Stuss, Murphy, Binns, & Alexander, 2003).  Executive cognitive functioning Based on the evidence above, it is clear that basic cognition plays a role in falls risk. However, there is one specific domain of cognition that appears to be most prominently linked to falls ? namely, executive cognitive functioning. Executive cognitive functions are our higher-level processing abilities, which require the maintenance and coordination of multiple cognitive subprocesses (e.g., Miyake, Friedman, Emerson, Witzki, & Howerter, 2000). According to Miyake and colleagues (2000), there are three main executive cognitive functions: mental set-shifting, information updating and monitoring, and inhibition of prepotent responses. Using factor analysis, the authors were able to determine that although the three functions are moderately correlated with each other, they do contribute differentially to performance on complex executive tasks.   Executive cognitive functions are subserved by the frontal lobe ? and more specifically, the prefrontal cortex (PFC) (e.g., Dempster, 1992). The frontal lobe is the most anterior portion of the brain, and comprises approximately one-third of the cerebral cortex in humans (e.g., Gazzaniga, Ivry, & Mangun,  5 1998). It is divided into three major components: 1) motor cortex, 2) premotor areas and the posterior portion of the cingulate cortex, and 3) PFC (e.g., Gazzaniga et al., 1998). The PFC is the most anterior portion of the frontal cortex, and is larger in humans relative to non-human animals. It is unique in that it is connected, via both afferent and efferent connections, with almost all other regions in the brain (e.g., Anderson, 2008; Gazzaniga et al., 1998). Due to this richly integrated relationship between the PFC and other brain regions, the PFC plays a fundamental role in numerous cognitive processes.  Several converging lines of evidence provide support for the notion that the PFC is critical for executive functions. First, damage to the PFC results in impaired executive functioning (e.g., Stuss & Alexander, 2000; Stuss, Gow, & Hetherington, 1992). The most well known case of prefrontal damage is Phineas Gage, who suffered from substantial injuries to his frontal lobe after a work-related accident forced an iron rod through his skull. Afterwards, he displayed a wide range of executive impairments such as poor planning, inhibition, and lack of responsibility (Harlow, 1868). Second, neuroimaging studies show increased activation in the PFC relative to other regions in the brain during tasks that engage executive functions (e.g., Baker et al., 1996). Third, structural maturation of the PFC mirrors development of executive function skills through childhood and adolescence (e.g., De Luca & Leventer, 2008). However, while it is clear that the frontal lobe and executive functions are linked, other regions in the brain work in consort with the PFC to maintain our higher order cognitive abilities (e.g., Anderson, 2008).  Evidence overwhelmingly supports the notion that impaired executive cognitive functioning is a hallmark characteristic of senior fallers. For example, Hausdorff et al. (2006) found that seniors with a history of falls performed worse on tests of executive function and attention relative to non-fallers, whereas global cognition scores were not different between the two groups. To ascertain the specific impairments that contribute to falls, several studies have employed laboratory-based paradigms to examine executive  6 cognitive functioning in senior fallers.  There are two specific executive cognitive functions that have been examined in relation to falls. First is response inhibition ? or our ability to suppress a prepotent response. In a sample of 658 seniors, Anstey et al. (2009) found that better accuracy/inhibition performance was associated with reduced falls risk, comparing non-fallers to both single and recurrent fallers. Similarly, Lord and Fitzpatrick (2001) reported poorer performance on the Stroop test in fallers relative to non-fallers. Along the same line, Springer et al. (2006) found that those with two or more falls in the past year performed significantly worse than non-fallers on both the Stroop and Go-NoGo ? another index of response inhibition.   A second component of executive cognitive functioning is planning, or set-shifting. This assesses the ability to switch thinking between different concepts. Within their neuropsychological battery, Lord and Fitzpatrick (2001) included the Trail Making B test, which requires participants to encircle alternating numbers and letters. The authors found that fallers performed significantly worse than non-fallers on this task, representing reduced set-shifting ability.   Dual-task performance  A second area of research concerning the relationship between cognition and falls is dual-task performance. Dual-tasking is the performance of two tasks concurrently (e.g., Baddeley, 1996; Kahneman, 1973). This often results in impaired performance on one or both tasks ? known as dual-task costs (DTC). Importantly, successful dual-task performance requires the employment of multiple cognitive and executive functions. Specifically, higher-level cognitive functions are required to attend to multiple stimuli at once and effectively allocate cognitive resources (e.g., Kahneman, 1973). This is evidenced by neuroimaging results demonstrating that the PFC is active during the performance of two tasks in conjunction, while each task  7 alone is not associated with such activation when performed in isolation (D'Esposito et al., 1995). Hence, dual-task paradigms offer insight into impairments in attentional processing, to the extent that otherwise intact cognitive performance may decline when paired with a secondary task.   While people of all ages incur DTC, performing two tasks concurrently is more difficult for older adults (e.g., Krampe, Schaefer, Lindenberger, & Baltes, 2011). Overall, it also appears that fallers have dual-task impairments greater than non-fallers, although the evidence has been somewhat equivocal. One of the first papers to emerge on falls and dual-tasks was the study investigating ?stops walking when talking? as a potential screening tool for falls (Lundin-Olsson, Nyberg, & Gustafson, 1997). The study was originally conceptualized based on the observation that frail elderly people often stop walking when they begin talking. The authors systematically observed participants and recorded whether or not they stopped walking when a conversation started and recorded falls prospectively over a six-month period. The authors found a significantly different distribution of falls between those who stopped walking and those who continued to walk while talking. Subsequently, Verghese and colleagues (2002) found that their test, ?walking while talking? (WWT) reliably predicted falls in seniors. Specifically, fallers were slower to walk while they concurrently recited letters of the alphabet (simple) or recited alternate letters of the alphabet (e.g., a,c,e; complex).   In contrast, some authors have found no additional value of dual-task performance compared with single-task performance for predicting falls. For example, Bootsma-van der Wiel et al. (2003) had participants walk while completing a verbal fluency task, which consisted of reciting as many names as possible (either animals or professions). Walking time, number of steps, and number of recited names were recorded. Among their sample, dual-task performance did not provide any additional information above and beyond single task performance in predicting falls over a one-year follow-up period.   8  Counter-intuitively, one study reported improved performance during dual-tasking in senior fallers (Beauchet et al., 2007). Participants were required to count backwards from 50 while walking 10 metres. Those who fell over the 12-month follow-up period tended to have superior performance in the dual-task compared with the single task. The authors attributed the dual-task benefits observed among fallers to their increased dependence on rhythm, which may have been facilitated by counting at a regular pace.  Several different cognitive tasks have been paired with mobility to determine the specific impairments that are observed among fallers. For example, in one study participants had to complete one of two separate cognitive tasks while walking ? participants had to either push a handheld button as quickly as possible in response to an auditory tone, or complete a visual-spatial decision task which required judging whether two clock-hands were on the same versus different sides of the clock (Faulkner et al., 2007). The researchers found that those with a history of falls walked significantly slower during the visual-spatial decision task, compared to non-fallers. Similarly, fallers experienced greater sway while completing a language processing task compared with a simple visual task (Shumway-Cook, Woollacott, Kerns, & Baldwin, 1997). Another study examined three separate dual-tasks, where participants walked while either simultaneously: 1. Listening to text; 2. Listening to text while completing a phoneme monitoring task; or 3. Completing subtraction problems (Springer et al., 2006). The authors reported increased swing time variability in fallers for all three dual-task conditions. Taken together, these results suggest that tasks requiring higher cognitive decision-making, rather than simple perceptual-motor tasks, are most related to falls.  Explaining why fallers have impaired performance on some dual-tasks but not others relates to models of a single cognitive resource pool (e.g., Kahneman, 1973) versus multiple resource pools (e.g., Navon & Gopher, 1979). Predicated on the theory that DTC result from cognitive demands outweighing resources  9 available, it is reasonable that easier tasks that do not tax the cognitive system would not interfere with gait and balance. For example, in the study by Beauchet and colleagues (2007), counting backwards is an easy, automatic task. That different secondary cognitive tasks result in different levels of impairment in fallers appears to favour a multiple resource model, where only tasks that tax the same pool of cognitive resources impair performance. However, definitive conclusions cannot be made based on the evidence provided in the literature thus far; future work linking dual-task impairments in fallers with models of cognitive resources are required.  Attention In my dissertation, I will examine how impaired attentional processing may be a risk factor for falls. Attention is essential for our ability to successfully navigate through the environment. This includes the ability to attend to relevant hazards or obstacles in our path. Briefly, what evidence is there that attention may be linked to falls? First, we know that attention is critical for visuo-motor abilities (e.g., Handy et al., 2005). Specifically, problems with visually-guided actions may manifest as impairments in planning and navigation ? and hence, falls. Second, impaired attentional processing is a hallmark characteristic in those with neurodegenerative disease (e.g., Sheridan & Hausdorff, 2007), such as Alzheimer?s, who are also two to three times more likely to experience a fall than cognitively healthy older adults (e.g., Tinetti et al., 1988). Third, lapses in attention are anecdotally cited as a leading cause for falls by seniors.   A few studies have been conducted to examine the relationship between attention and falls. In a cross-sectional study comparing those with a history of falls to non-fallers, performance on tasks which employed speed/executive attention was significantly related to falls (Holtzer et al., 2007). Along the same line, fallers have impaired performance on the Digit Symbol Test, which also assess processing speed and attention (Lord & Fitzpatrick, 2001). Another study comparing fallers and non-fallers found significant differences  10 between groups on a selective attention task (Woolley et al., 1997). Specifically, in a visual search task, participants with a recent history of falls had poorer accuracy and RTs for locating targets amongst a display of distractors compared with those without a history of falls. Rosano and colleagues (2008) found that gait characteristics linked to increased falls risk, such as shorter steps and longer double support times, were significantly associated with smaller gray matter in brain regions responsible for motor, visual-spatial orientation, and processing speed/executive control functions. Such neuroimaging evidence supports the notion that cognitive and motor processes are functionally connected in the brain.  Is attention an executive cognitive function? Attention is an umbrella term, encompassing many distinct forms. Thus, classifying attention as an executive cognitive function may depend on the type of attention that is being referred to. For instance, reflexive visual spatial attention, which is processed in a bottom-up manner, may not be considered ?executive?; attention in this instance is not consciously controlled. In other cases, such as volitional visual spatial attention, attention can be considered as an executive cognitive function. This is based on the fact that both share similar anatomical locations in the brain ? namely, prefrontal cortex, and that this form of attention requires the conscious, deliberate decision of where to orient attention. For our purposes, the forms of attention discussed throughout this dissertation can be considered as an ?executive cognitive function?.  Aspects of attention Generally, attention refers to the selection and focus of particular information while ignoring other less salient or relevant information (e.g., James, 1890). Attention, however, is a broad construct with several sub-classes. Three aspects of attention ? visual-spatial, divided, mind-wandering ? will be discussed extensively in this dissertation. Thus, I will present a brief overview distinguishing between these aspects of  11 attention and summarize relevant research providing the impetus for why they may be linked to falls.  Visual-spatial attention Visual-spatial attention can be considered a sub-type of selective attention. Broadly, selective attention refers to our ability to actively attend to certain information (objects or location) while ignoring task-irrelevant information, and requires executive attentional control (e.g., Treisman, 1969). We tend to attend to the most salient information, while ignoring less important or interesting environmental stimuli. Selective attention can occur in different sensory modalities, such as visual or auditory. In addition, different types or levels of information can be selected. For example, in visual selective attention, one may select whole objects or distinct features of objects (e.g., colour or shape). Selective attention can be studied in the laboratory using tasks that require participants to focus on one stimuli, while ignoring others, such as a dichotic listening task for auditory attention (e.g., Treisman, 1960) or the Eriksen Flanker task for visual attention (Eriksen & Eriksen, 1974).  Visual-spatial attention, specifically, is the selection of visual information in a particular spatial location (e.g., Posner, 1980). Visual-spatial attention can be further broken down according to whether it is volitional/controlled or reflexive/automatic (e.g., Corbetta & Shulman, 2002). In volitional visual-spatial attention, the selection of information is consciously decided in a top-down manner, usually consistent with behavioural goals. In contrast, reflexive visual-spatial attention involves the automatic drawing of attention through bottom-up processes, such as seeing a bright flash in the visual periphery. Importantly, visual-spatial attention is commonly described as a ?spotlight,? where attention is focused in one area of visual space, with heightened perceptual processing of stimuli falling within the spotlight (e.g., Posner, 1980). Visual-spatial attention is typically assessed behaviourally and electrophysiologically, where both reaction times and neural activity in visual cortex are enhanced for stimuli appearing in an attended location (e.g.,  12 Mangun, Hillyard, & Luck, 1993; Posner, 1980).  Previous research that I conducted examined differences between senior fallers and non-fallers in their ability to orient visual-spatial attention (Nagamatsu, Carolan, Liu-Ambrose, & Handy, 2009). In this study, I used a visual-spatial cueing paradigm to direct attention to either the left or right side of visual space (Posner, 1980). I employed event-related potentials (ERPs), a neuroimaging technique which measures electrical activity in the brain via electrodes on the scalp. Compared with non-fallers, fallers exhibited significantly reduced amplitude for the P1 component, which reflects sensory processing in visual cortex and modulates with attention (e.g., Handy & Mangun, 2000; Luck et al., 1994; Mangun & Hillyard, 1991). This finding, however, was specific the left visual field. In other words, fallers had reduced sensory/perceptual responses to stimuli appearing on the left. What has remained unknown thus far ? and is of central importance to falls ? is the extent to which visual-spatial attention to task irrelevant information might be altered as a function of falls risk.  Divided attention Divided attention ? commonly known as dual-tasking ? is our ability to attend to two or more tasks simultaneously (e.g., Baddeley, 1996; Kahneman, 1973). Importantly, performing multiple tasks concurrently commonly interferes with performance on both tasks. Such interference is attributed to both structural and capacity limitations (e.g., Kahneman, 1973). Furthermore, successful task performance (on one or both tasks) is dependent on the effective allocation of attention resources (e.g., Kahneman, 1973). In this context, divided attention can be considered as a superordinate form of attention in contrast to selective attention, as described above. In short, each task within a dual-task paradigm requires its own unique attentional set and selection of relevant information.    13 Literature on falls and dual-tasking has been presented in detail above. What is currently unknown in this field of research, however, is how dual-task findings in fallers might apply to real-world situations. Specifically, commonly used dual-task paradigms in falls research are laboratory based and not ecologically valid. Furthermore, most studies on dual task performance in fallers measure the physical consequences of dual-tasking, such as reduced balance or altered gait. My primary interest, therefore, is examining how dual-tasking might impact cognitive processes in fallers, such as decision making/judgments ? and how these extend to real-life situations.  Mind-wandering Mind-wandering is our natural tendency for our thoughts to drift off task. In the context of attention, mind-wandering refers to what we pay attention to: external versus internal information. Importantly, the occurrence of mind-wandering has been attributed to lapses in attention control ? or the inability to focus on the central task at hand while ignoring internal distractions (e.g., Smallwood, in press). In the laboratory, mind-wandering is often studied using ?experience sampling?, which asks participants at frequent but unpredictable intervals to self-report their current attentional state while they are simultaneously engaged in a mundane, repetitive cognitive task (e.g., Kam et al., 2011). Electrophysiological and/or behavioural output can then be compared during periods where participants were on-task relative to mind-wandering. In this manner, the neurocognitive disruptions that result from drifting off-task can be examined. Further, this allows for the frequency of mind-wandering events to be recorded as well.  There has been no research to date on the relationship between falls and mind-wandering; therefore, we can only speculate what relationship might be present between the two. Based on our current knowledge that fallers exhibit reduced response inhibition and that mind-wandering may be due to a lack of attentional control, fallers would be expected to have an increased propensity to mind-wander. Furthermore, mind- 14 wandering might be considered as a form of ?dual-tasking?, with two ?tasks? being performed concurrently (i.e., the primary task plus engaging in task-irrelevant thoughts). Given that falls are associated with impaired dual-task performance, I hypothesize that fallers would also show reduced cognitive performance during periods of mind-wandering.  Overview of Dissertation My dissertation will center on the hypothesis that impaired attentional processing is a risk factor for falls in older adults. Throughout this dissertation, I will provide evidence that impaired attentional processing alters how environmental stimuli are processed as a function of falls risk. The overarching significance of examining the relationship between attention and falls risk is that it will help elucidate the underlying mechanisms that lead to falls, which may ultimately result in better screening and prevention strategies.  Main Research Questions The aim of my dissertation is to answer the following three research questions: 1. What specific types of attention are associated with falls/falls risk? Specifically, I examine how three distinct types of attention are altered as a function of falls/falls risk: 1) Visual-spatial attention; 2) Divided attention; and 3) Mind-wandering. 2. How might impaired attentional processing contribute to increased falls risk? 3. What underlying neural structures are implicated in the relationship between attention and falls risk?  Methodology Below is a brief summary of how constructs used throughout my dissertation are measured/assessed.    15 Measures of falls and falls risk  Falls and falls risk are two related but dissociable constructs. For studies included in my dissertation that compare fallers and non-fallers, falls status was based on history of falls gathered from monthly falls calendars. This is the most reliable method to report falls history, given that recall can be subject to retrospective memory bias. For studies that instead examine ?falls-risk?, risk was determined based on physiological measures established to predict falls, such as the Physiological Profile Assessment (PPA) ? a valid and reliable measure of falls risk based on a composite score from five independent measures (hand reaction time, contrast sensitivity, proprioception, leg extension strength, and sway) (Lord, Menz, & Tiedemann, 2003). A PPA z-score of 0-1 indicates mild risk of falls, 1-2 indicates moderate risk, 2-3 indicates high risk, and 3 and above indicates marked risk (Lord et al., 2003). The PPA has been shown to have 75% predictive accuracy for falls in older adults (Lord et al., 2003).  Measures of attention My research focuses on identifying both the behavioural and the neural underpinnings of cognitive impairment as a function of falls risk. Thus, my methodological approach is to examine behavioural performance and corresponding functional differences in the brain.  Behavioural Behavioural measures provide an object way to assess performance. The behavioural measures that I focus on are reaction times (e.g., button presses to visual targets) and accuracy (e.g., how many targets do they correctly respond to).  Neuroimaging Event-related potentials (ERPs): ERPs offer a non-invasive approach to measuring time-locked electrical  16 activity in the brain, allowing us to infer the level of sensory and/or cognitive processing that is occurring in response to an environmental stimulus. By comparing the response to different classes of stimuli, we can understand how different conditions modulate sensory and cognitive processing. With the advantage of high temporal resolution, ERPs also reveal the underlying time course of processing (e.g., early versus late).  Functional magnetic resonance imaging (fMRI): fMRI is a neuroimaging technique that measures changes in levels of oxygenated blood to infer cognitive processing at a particular location in the brain. Contrasting with ERPs, fMRI provides high spatial resolution, but reduced temporal resolution, due to the time lag inherent to the hemodynamic response. Thus, the two neuroimaging techniques are complementary to each other, with each aimed at answering a different type of research question. Both neuroimaging techniques are used in my dissertation.  Overview of Studies To address the above research questions, my dissertation is comprised of four studies (each presented in a separate chapter), summarized here.  In Chapter 2, I examine how altered attention to task-irrelevant probes is associated with falls risk in older adults. Participants completed a target discrimination task at fixation while peripheral probes were presented on the right and left side of visual space. Using ERPs, we assessed sensory processing to the probes and found that reduced N1 amplitude to probes on the left side and measured at ipsilateral electrode sites was significantly associated with increased falls risk.  The study presented in Chapter 3 was designed to assess the extent to which the performance of a cognitively difficult task would impair decision-making and mobility performance during a virtual street-crossing task. I found that older adults at-risk for  17 falls made poorer judgments and exhibited impaired performance during a dual-task condition where they were required to simultaneously engage in a telephone conversation. In Chapter 4, I investigate whether falls are associated with frequency of mind-wandering and/or altered neurocognitive consequences of being off task. This was accomplished using behavioural and neuroimaging measures to assess cognitive functioning during a sustained attention task. I found that falls were significantly associated with an increased propensity to mind-wander. In contrast, falls were not associated with changes in sensory or cognitive processing during periods of mind-wandering. Lastly, Chapter 5 examines whether selective attention might be associated with falls risk. Furthermore, this chapter aims to examine what functional brain regions could be the arbiter between altered attention and falls risk. In this study, fMRI was used to examine the functional activation during a test requiring selective attention and response inhibition. I found that reduced physiological falls risk was negatively associated with functional activation in the left orbital frontal cortex extending towards the insula (OFC-In) and positively associated with activation in the right paracingulate gyrus extending towards the anterior cingulate cortex (PCG-ACC).  Overall, the research presented in this dissertation provide a coherent theory that augments the existing literature on executive cognitive functioning and falls, and implicates altered attentional processing as a key risk factor for falls. Following the presentation of the four studies outlined above, my dissertation concludes with an integrated discussion of what we have learned about the relationship between attention and falls risk, a revisit of the main research questions presented above, and provide an overview of limitations and future directions in this relatively new but rapidly expanding line of research.    18 CHAPTER 2: Altered attention to task-irrelevant stimuli is associated with falls risk in seniors Introduction It is well known that executive cognitive functions, or higher-level processes including response inhibition, set-shifting, and working memory, play a critical role in mobility and balance (e.g., Lord et al., 2007). This relationship is illustrated in older adults who experience falls ? a pressing health care issue for our aging society. Fallers have reduced global cognitive functioning (e.g., Tinetti et al., 1988) and perform worse in laboratory tests of higher-level processing (for a review see Hsu, Nagamatsu, Davis, & Liu-Ambrose, 2012). One executive function that is essential for successful navigation through the environment is visual-spatial attention (e.g., Albert, Reinitz, Beusmans, & Gopal, 1999). For instance, visual-spatial attention during a volitional orienting task was previously found to be altered in older adults with a history of falls relative to those without such a history (Nagamatsu et al., 2009). Although differences in sensory processing afforded at an attended location ? known as attentional facilitaton ? were observed between fallers and non-fallers in the above study, it is currently unknown whether similar alterations in visual-spatial attention may exist for the implicit processing of task-irrelevant information.  Why might we expect that visual-spatial attention to task-irrelevant information would be associated with falls? First, the capacity to process external information depends on the availability of excess resources after the allocation of sufficient attention to the primary task. According to the capacity model of attention (e.g., Kahneman, 1973), we have a finite reserve of cognitive resources; the more resources a given task requires, the fewer that are available to distribute to a secondary task. This model is supported by previous studies that have demonstrated that early sensory processing of peripheral stimuli varies according to the perceptual difficulty of the task at fixation (e.g., Handy, Soltani, & Mangun, 2001). Specifically, the P1 ERP  19 component, which indexes early sensory processing and modulates as a function of attention, had reduced amplitude in response to task-irrelevant stimuli when perceptual load of the primary task was increased. Senior fallers have been proposed to possess reduced cognitive resources, as evidenced by their impaired ability to perform two tasks concurrently (i.e., dual-task) (e.g., Faulkner et al., 2007; Lundin-Olsson et al., 1997; Shumway-Cook et al., 1997; Springer et al., 2006; Verghese et al., 2002). These findings suggest that fallers may not have adequate cognitive resources to dedicate to the sensory or cognitive processing of task-irrelevant information while attending to a central task.  Second, recent evidence suggests that fallers may have a ?narrowed? focus of attention (Liu-Ambrose, Nagamatsu, Leghari, & Handy, 2008) ? which may reduce their ability to process peripheral stimuli. Attention is often described as a spotlight, where information appearing within the spotlight is preferentially processed at a sensory (and subsequently cognitive) level, relative to information outside the spotlight (Posner, 1980). Evidence that fallers may have a narrowed spotlight of attention comes from a cross-sectional study, where fallers demonstrated superior behavioural performance compared with non-fallers on the Eriksen Flanker task (Liu-Ambrose, Nagamatsu, et al., 2008), an executive task that assesses selective attention and conflict resolution (Eriksen & Eriksen, 1974). In this version of the Eriksen Flanker task, participants were required to indicate the direction that a central arrow was pointing while ignoring flanking arrows on either side of the target; trials were either congruent, with all arrows pointing in the same direction (e.g., <<<<<), or incongruent, with the central arrow pointing in the opposite direction as the distractor arrows (e.g., <<><<). Performance on this task relates to the extent to which one can ignore the task-irrelevant distractors surrounding the target. We reported that participants with a recent history of falls demonstrated less interference from incongruent trials relative to those without a history of falls (Liu-Ambrose, Nagamatsu, et al., 2008). These counterintuitive results suggest that fallers may be better able to respond to central stimuli without interfering input from peripheral non-targets.   20 Taken together, the evidence provided above suggests that fallers may have reduced visual-spatial attention to task-irrelevant information. With approximately 30% of community dwelling seniors experiencing one or more falls per year, identifying the key cognitive contributors to falls risk will aid with the development of effective treatment and prevention strategies for reducing falls. Importantly, the neurocognitive risk factors for falls are not currently well understood. Further, elucidating the exact visual-spatial impairments exhibited by fallers will advance our understanding of the potential underlying mechanisms that link cognitive functioning and falls. Hence, the primary aim of our current study was to examine whether visual-spatial attention to task-irrelevant stimuli is associated with falls risk in older adults.  The secondary aim of our study was to evaluate whether visual field asymmetries are associated with falls risk. We previously found that older adults with a history of falls exhibited reduced attentional facilitation during an orienting task ? but that this effect was specific to the left visual field (Nagamatsu et al., 2009). Given the contralateral nature of visual processing, impaired visual-spatial attention exclusively on the left side of visual space is indicative of problems with function and/or structure of the right hemisphere. To the point, research on split-brain patients has revealed that the right hemisphere orients attention to both visual fields, whereas the left hemisphere only orients to the right side of space (e.g., Mangun et al., 1994). Hence, reduced function in the right hemisphere would result in a lack of attention to the left side of space, without adequate compensation from the left hemisphere. Thus, comparing the level of visual-spatial attention between the two visual hemifields may provide insight into the underlying neurocognitive link between attention and falls risk.   Towards addressing our specific study aims, we conducted a cross-sectional study of older adults, examining the relationship between visual-spatial attention and falls risk. Participants completed a visual target discrimination task at fixation while task-irrelevant probes were presented in the left and right  21 periphery. Task complexity was manipulated by varying difficulty of the primary task at fixation. To ascertain the relationship between visual-spatial attention to task-irrelevant probes and falls risk, we measured ERPs during task performance. Accordingly, the ERP components of interest in our study included the P1 and N1, both of which index early sensory processing ? and consequently, attentional facilitation (e.g., Mangun et al., 1993). Key to examining visual-spatial attention in this study, these components have increased amplitudes to stimuli appearing at an attended location relative to those in an unattended location (e.g., Mangun et al., 1993). This modulation in amplitude provides us with the opportunity to infer where in visual space participants are attending at any given time. Importantly, the P1 and N1 components are functionally dissociable; whereas the P1 represents facilitation of early sensory processing, the N1 has been implicated in attentional orienting to task-relevant information and discriminative processing (e.g., Luck, Heinze, Mangun, & Hillyard, 1990; Vogel & Luck, 2000). Additionally we examined performance on the target discrimination task via reaction times, accuracy, and the P300 component. Briefly, the P300 component represents cognitive evaluation of the stimuli and is required for the maintenance of working memory (e.g., Polich, 1996). Falls risk was determined using the PPA ? a widely used measure that accounts for multiple physiological risk factors for falls (Lord et al., 2003). Based on our previous findings, we hypothesized that falls risk would be associated with reduced attentional processing to peripheral probes, and that this effect would be further characterized by hemispheric asymmetries.  Methods Participants Our study included thirty-one participants (20 female). The mean age was 75.23 (SD = 3.43) years. All participants had normal or corrected-to-normal vision and 27 participants were right-handed. Participants had intact cognitive functioning, as indicated by a mean MMSE score of 28.39 (SD = 1.58). Participants in our study were part of a large-scale cross-sectional study (CogMob) comparing cognitive profiles of senior  22 fallers and non-fallers using fMRI. They were recruited via advertisements in local newspapers. Upon consenting to participation in the larger study, participants were individually asked if they would be interested in completing a second, independent, sub-study. Those that agreed to be contacted were then scheduled to participate in this secondary study.  Falls risk We assessed physiological falls risk using the PPA, which is a valid and reliable measure of falls risk (Lord et al., 2003) (Prince of Wales Medical Research Institute, AUS). The PPA is a z-score based on five separate physiological measures (hand reaction time, contrast sensitivity, proprioception, leg extension strength, and sway) to indicate relative risk of falls (Lord et al., 2003). A PPA z-score of 0-1 indicates mild risk of falls, 1-2 indicates moderate risk, 2-3 indicates high risk, and 3 and above indicates marked risk (Lord et al., 2003). The PPA has been shown to have 75% predictive accuracy for falls in older adults (Lord et al., 2003).  Descriptive measures Descriptive measures were collected from all participants during the CogMob study. Demographic information was ascertained via questionnaire. Falls history over the past 12 months was reported based on subjective recall and corroborated by an immediate family memory or close friend. General cognitive functioning was assessed using the Montreal Cognitive Assessment (MoCA) (Nasreddine et al., 2005). We assessed depression using the Geriatric Depression Scale (GDS) (Yesavage, 1988) and number of comorbidities using the Functional Comorbidity Index (FCI) (Groll, To, Bombardier, & al., 2005). Balance confidence was assessed using the Activities-Specific Balance Confidence scale (ABC) (Powell & Myers, 1995). The Timed Up and Go test (TUG) (Podsiadlo & Richardson, 1991) was used to assess general balance and mobility. For this task, participants start from a seated position and are instructed to stand up,  23 walk three metres, return to their chair, and sit back down. The average time to complete the TUG on two separate trials was recorded, with faster times indicating better performance.  Procedure Trial sequence and timing are provided in Figure 2.1. Stimuli were presented on an 18 inch colour monitor approximately 100 cm from each participant. In this experiment, participants completed a visual target discrimination task at central fixation under each of two conditions: low load and high load. Each trial began with a fixation cross in the centre of the screen for 500-700 msec. Next, one of six possible stimuli appeared, with equiprobability. Targets/non-targets were presented at the centre of the screen and consisted of a bar that randomly varied by colour (purple or green) and orientation (horizontal or vertical); each configuration of the bars occurred an equal number of times within each block. Task-irrelevant probes included three by three black and white checkerboards presented on either the left or right side of the screen. Blocks consisted of 96 trials, with 16 of each trial types occurring within each block. Participants were instructed to make a manual response on a game pad (using either their left or right index finger, counterbalanced between participants) as quickly and accurately as possible if the bar at fixation matched the criteria provided to them at the beginning of the session. For the low load condition, participants were instructed to respond if the bar was a pre-specified colour (e.g., green; counterbalanced between participants), regardless of bar orientation. For the high load condition, participants were told to respond if the bar matched one of two particular colour and orientation combinations (e.g., either green and vertical or purple and horizontal; counterbalanced between participants). Participants were asked to ignore the checkerboards on the left and right side of the screen because they were tangential to the task. All participants completed eight consecutive blocks of each condition (i.e., low and high load); the order that the conditions were completed was counterbalanced between participants.   24  Figure 2.1. Stimulus presentation and timing.   Electrophysiological recording and analysis During task performance, electroencephalograms (EEGs) were recorded from 32 active electrodes (Bio-Semi Active 2 system) evenly distributed over the head. All EEG activity was recorded relative to two scalp electrodes located over medial-frontal cortex (CMS/DRL), which served as the ground, and the average of the two mastoids (left/right), which served as the reference. Data was recorded using a second order low pass filter of 0.05 Hz, with a gain of 0.5, and digitized on-line at a sampling rate of 256 samples-per-second. To ensure proper eye fixation and allow for the correction and/or removal of events associated with eye movement artifacts, vertical and horizontal electro-oculograms (EOGs) were also recorded, the vertical EOG from an electrode inferior to the right eye, and the horizontal EOG from an electrode on the right outer canthus.    25 Electrophysiological analysis was performed using ERPlab (http://erpinfo.org/erplab/), a Matlab package used in conjunction with EEGLAB (Delorme & Makeig, 2004). Continuous data was first separated into  -1500 to 1500 msec epochs time-locked to stimulus presentation, then grouped into bins according to trial type. Artifact rejection was used to eliminate epochs during which detectable eye movements, blinks, or muscle potentials occurred. For artifact rejection, the moving windows peak-to-peak option within ERPlab was implemented, with thresholds customized for each participant. Data was then low- and high-pass filtered using the IIR Buttersworth option in ERPlab, at 30 and 0.1 Hz, respectively. Statistical quantification of ERP data was based on mean amplitude measures relative to a ?200 to 0 pre-stimulus baseline. Electrode sites and time windows for our ERP analyses were chosen based on the extant literature for these well-studied components (see below).   Data analysis Behavioural performance measures comprised of accuracy and reaction times. Accuracy was assessed using d-prime, a measure of signal detection that corrects for response bias. Larger d-prime values represent better performance. Reaction times were measured for correct trials only. Faster reaction times represent better performance. Behavioural data was analyzed using paired t-tests to compare performance during high versus low load blocks. Our electrophysiological data was analyzed using repeated-measures ANOVAs to compare mean amplitudes of the ERP components as a function of load (high vs. low), trial type (left vs. right visual field for the P1 and N1; target vs. non-target for the P300), and electrode location. Correlations between key variables of interest were computed using Pearson?s product-moment coefficient. All data was analyzed using SPSS (Version 20 for MAC), with alpha set at p < 0.05.     26 Results Descriptive measures and falls risk Descriptive data are reported in Table 2.1. The mean PPA score was 0.45 (SD = 0.95), indicating an overall low risk of falls. Falls risk ranged from low risk to high risk, as exhibited by the range for PPA scores from -1.29 to 2.80. PPA scores were significantly correlated with number of falls over the past 12 months, r(31) = 0.51, p = 0.003, balance confidence (ABC score), r(31) = -0.42, p = 0.02, and TUG time, r(31) = 0.47, p = 0.008. Correlations between PPA score and other descriptive measures were non-significant (all p?s > 0.05).     27 Table 2.1. Descriptive information.  Variables1 n = 31 Age, years 75.23 (3.43) Falls history, No. 2.42 (3.60) Education, No. (%):  Less than grade 9 1 (3.2) Grade 9-13 without certificate/diploma 2 (6.5) High school certificate/diploma 0 (0.0) Trades or professional certificate/diploma 7 (22.6) Some University without certificate/diploma 2 (6.5) University certificate/diploma 4 (12.9) University degree 15 (48.4) MoCA2 24.77 (2.97) MMSE3 28.39 (1.58) Timed Up and Go (TUG) 8.63 (3.58) Geriatric Depression Scale (GDS) 0.57 (1.01) Physiological Falls Risk (PPA) 0.45 (0.95) Comorbidities ABC4 3.06 (2.10) 84.68 (16.50) 1Data presented as mean (SD), unless otherwise indicated. 2MoCA=Montreal Cognitive Assessment; Maximum 30 points. 3MMSE=Mini-Mental Status Examination; Maximum 30 points. 4ABC=Activites-Specific Balance Confidence scale; Maximum 100 points.    28 Behaviour Behavioural results are presented in Table 2.2. One participant did not complete both experimental conditions, and was therefore excluded from analyses with ?low load? as a factor. Overall, participants were faster to respond and had higher accuracy during low load blocks relative to high load. This was revealed via significant differences in performance between block types, t(29) = 13.55, p < 0.001 and t(29) = 4.85, p < 0.001, for reaction times and d-prime respectively.  Table 2.2. Behavioural results.    High load (n = 31)  Low load (n = 30)    Reaction time (msec) 0.5505 (0.0798)  0.4032 (0.0598) Accuracy (d?) 2.98 (0.96)  4.19 (0.79)  Examining the correlations between our behavioural measures and falls risk (PPA score), we found that both reaction time and accuracy were significantly associated with falls risk ? but only in the high load condition. Specifically, in high load conditions, better performance (i.e., faster reaction times and higher accuracy) was correlated with lower falls risk. The correlation between reaction time and PPA in high load condition was r(31) = -0.37, p = 0.04. The correlation between d-prime and PPA in the high load condition was r(31) = -0.47, p < 0.01. In contrast, the correlations between behavioural performance during the low load blocks and falls risk were not significant, p = 0.63 for reaction time and p = 0.73 for d-prime.     29 Electrophysiology Six participants were excluded from the electrophysiological analysis due to excessive noise in their data (i.e., large pre-stimulus baseline amplitudes or high artifact rejection rates).   P1 ERP component We measured the P1 ERP component time-locked to peripheral probes to assess attentional facilitation to task-irrelevant stimuli. Grand-averaged waveforms for the P1 component are presented in Figure 2.2 and mean amplitudes are provided in Table 2.3. In our analysis, the electrode sites we used were OL+ and OR+, which are the average of lateral occipital-parietal sites (P3, P7, PO3, and O1 for OL+; P4, P8, PO4, and O2 for OR+) (e.g., Mangun et al., 1993). To accurately capture the P1 component, for each electrode site and condition, the time windows of analysis were centered on the peak amplitudes of the component in the grand-averaged waveform. For the P1, the windows were 130-170 msec post-stimulus for ipsilateral sites and 110-150 msec for contralateral sites.      30 Figure 2.2. ERP waveforms for the P1 and N1 components.       31 Table 2.3. Mean amplitudes for P1 and N1 ERP components time-locked to task-irrelevant probes.    High load (n = 25) Low load (n = 25) P1a       Left ipsilateral (OL+) 1.17 (0.94) 1.19 (1.13)     Left contralateral (OR+) 0.93 (0.99) 0.67 (1.42)     Right ipsilateral (OR+) 1.24 (1.34) 0.93 (1.22)     Right contralateral (OL+) 1.11 (1.32) 0.83 (0.97) N1       Left ipsilateral (OL+) 0.33 (1.87) 0.35 (1.72)     Left contralateral (OR+) -1.17 (2.00) -0.89 (2.19)     Right ipsilateral (OR+) 0.24 (1.93) 0.04 (2.22)     Right contralateral (OL+) -1.29 (2.36) -1.57 (2.49) aMean amplitudes measured in uV.  For the P1 ERP component, there were no significant effects of load, visual field, or laterality, as revealed by the repeated measures ANOVA (all p?s > 0.12). We examined the association between attentional facilitation and falls risk by calculating the correlations between mean amplitudes of the P1 component and PPA scores. Because there were no significant modulations of load in the repeated measures ANOVAs, we collapsed high and low load conditions. Therefore, we had four measures of interest for each component: Left and right visual probes, each measured at ipsilateral and contralateral electrode sites. There were no significant associations between P1 amplitude and PPA (all p?s > 0.22).    32 N1 ERP component The N1 ERP component was measured time-locked to peripheral probes to assess attentional facilitation to task-irrelevant stimuli. Grand-averaged waveforms for the N1 component are presented in Figure 2.2 and mean amplitudes are provided in Table 2.3. The same lateral occipital-parietal electrode sites were used as for the P1: OL+ and OR+. The time windows of analysis were centered on the peak amplitudes of the component in the grand-averaged waveform. For the N1, the windows were 180-220 msec for ipsilateral sites and 160-200 msec for contralateral sites.  For the N1 ERP component, amplitudes measured at electrode sites contralateral to peripheral probes were larger than at ipsilateral sites, as demonstrated via a significant main effect of laterality, F(1,24) = 34.86, p < 0.001. The main effects of load and visual field were non-significant (p?s > 0.24). The same method for the P1 was applied to the N1 for our correlation calculations, where the factor of load was collapsed. As can be observed in Figure 2.3, we found that a larger N1 amplitude (i.e., more negative) was associated with lower falls risk. This was confirmed via a significant correlation between falls risk and mean amplitude of the N1 for peripheral probes presented in the left visual field, measured at ipsilateral electrode sites, r(25) = 0.44, p < 0.03.      33 Figure 2.3 Correlations between mean ERP amplitudes (uV) and falls risk for the N1 for stimuli presented in the left visual field, measured at ipsilateral sites (OL+).      P300 ERP component We assessed cognitive processing to task-relevant stimuli via the P300 ERP component time-locked to central targets and non-targets. Grand-averaged waveforms for the P300 ERP component are presented in Figure 2.4 and mean amplitudes are provided in Table 2.4. In our analysis, we used midline electrode sites (CZ, PZ, OZ) at a time window of 430-530 msec post-stimulus, centered around the peak of the P300 in the grandaveraged waveform (e.g., Eimer, 1996, 1998). Trials were separated into those with ?Go? targets and those with ?NoGo? non-targets. Our analysis only included those trials with correct responses.          ?3??2??1?0?1?2?3?4?5??1.5? ?1? ?0.5? 0? 0.5? 1? 1.5? 2? 2.5? 3? 3.5?Mean?ERP?amplitude?Falls?risk?(PPA?score)? 34 Figure 2.4. ERP waveforms for the P300 component.    35 Table 2.4. Mean amplitudes for P300 ERP component time-locked to central stimuli   High load (n = 25) Low load (n = 25) Targetsa       CZ 6.14 (4.96) 8.80 (5.20)     PZ 7.50 (5.09) 11.28 (4.22)     OZ 3.85 (4.22) 6.96 (3.75) Non-targets       CZ 5.09 (4.20) 7.96 (4.30)     PZ 5.54 (4.14) 7.04 (3.58)     OZ 2.25 (3.40) 3.64 (3.32) aMean amplitudes measured in uV.  Based on our Figure 2.4, the P300 amplitude was largest at the PZ electrode site, and larger during low load blocks compared to high load, and for targets relative to non-targets. Indeed, this is what we found with main effects of load F(1,24) = 16.33, p < 0.001, target type F(1,24) = 15.14, p = 0.001, and electrode location F(2,48) = 33.08, p < 0.001. Furthermore, there was a significant load x target type x electrode interaction F(2,48) = 4.94, p = 0.01. For our correlational analysis, larger positive deflections in P300 amplitude during low load blocks were associated with higher risk of falling (see Figure 2.5). In particular, there was a significant correlation between the P300 amplitude measured at electrode site PZ and PPA score, r(25) = 0.43, p = 0.03. All other correlations were non-significant (p?s > 0.07).     36 Figure 2.5 Correlations between mean ERP amplitudes (uV) and falls risk for the P300 for targets presented at fixation during low load blocks.    Control analyses To ensure that our significant correlations presented above were not caused by extreme outliers, we converted scores on the three relevant variables (PPA, N1 amplitude to left probes measured at ipsilateral sites, and P300 amplitude to targets during low load blocks) to z-scores. This revealed that there were no extreme scores (all z-scores < 2.5). Furthermore, our variables were normally distributed, as indicated by both skewness and kurtosis values < 1.5 for all three variables.  Discussion Our present study was aimed at examining the association between visual-spatial attention to task-irrelevant stimuli and falls risk in older adults. To that end, we report two main results. The first is that falls risk was associated with reduced attentional facilitation to task-irrelevant stimuli presented in the left visual field ? but this relationship was only evident in the N1 component measured at ipsilateral electrode sites.  0?5?10?15?20?25??1.5? ?1? ?0.5? 0? 0.5? 1? 1.5? 2? 2.5? 3? 3.5?Mean?ERP?amplitude?Falls?risk?(PPA?score)? 37 The second result pertains to performance and processing of the task-relevant information. Specifically, increased falls risk was significantly related to poorer behavioural performance during the visual discrimination task under high load conditions. Moreover, increased falls risk was also associated with an increased level of cognitive processing for targets during low load conditions. Our results support current prevailing theories on the relationship between executive cognitive functions and falls risk, and provide insight into the specific impairments that may contribute to falls.  Primary finding Our primary result regarding the relationship between falls risk and attentional facilitation corroborates and extends our previous finding that older adults with a history of falls have altered visual-spatial attention to the left hemifield (Nagamatsu et al., 2009). Based on the notion that the right hemisphere is exclusively responsible for orienting attention to the left side of visual space (e.g., Mangun et al., 1994), this pattern of results is consistent with what is observed in patients with left visual neglect following right hemisphere damage (e.g., Bublak, Redel, & Finke, 2006; Reuter-Lorenz, Kinsbourne, & Moscovitch, 1990). Hence, our results suggest that falling in older adults might stem from a mild form of visual neglect.  Given that reduced attentional facilitation was specifically observed at ipsilateral electrode sites, there are at least two potential underlying factors. First falls risk may be associated with impaired inter-hemispheric transfer of visual information across the corpus callosum. The corpus callosum is a dense bundle of neural fibres that connects the two cerebral hemispheres of the brain. Previous studies have reported a critical link between white matter integrity in the corpus callosum and both cognitive and mobility measures (e.g., Bhadelia et al., 2009; Frederiksen et al., 2011; Moscufo et al., 2011; Moscufo et al., 2012; Ryberg et al., 2011; Ryberg et al., 2007). Of particular relevance, atrophy in the splenium ? a posterior region of the corpus callosum ? appears to be most associated with reduced general mobility (e.g., Frederiksen et al.,  38 2011; Moscufo et al., 2011; Moscufo et al., 2012) as measured by the Short Physical Performance Battery (SPPB) (Guralnik, Ferrucci, Simonsick, Salive, & Wallace, 1995). Importantly, the splenium is implicated in inter-hemispheric transfer of visual and somatosensory information from occipital and posterior-inferior parietal cortices (e.g., Park et al., 2008); this information is fundamental for the integration of visual and spatial inputs to motor responses (e.g., Moscufo et al., 2011). Within this context, our data suggests that reduced attentional facilitation may be a functional consequence of atrophy in the posterior corpus callosum in those at-risk for falls.   The second potential interpretation for the observed reduction in attentional facilitation at ipsilateral electrode sites in our study is that the posterior right hemisphere may have problems with generating a signal to transfer visual information to the left hemisphere. For example, unilateral neglect is most often the result of damage to the right posterior parietal lobe ? and more specifically, the inferior parietal lobule or the temporo-parietal junction (e.g., Halligan, Fink, Marshall, & Vallar, 2003). This anatomical identification of the neural basis of neglect has been corroborated by transcranial magnetic stimulation (TMS) studies (e.g., Halligan et al., 2003). Regardless of the underlying mechanism responsible for the pattern of results we have obtained, our data fits with a neglect model of hemispheric differences in attention to visual space. We note that these two possible explanations are not mutually exclusive; however, elucidating the role of functional and/or structural neural correlates in the relationship between visual-spatial attention and falls risk is required in future research.  Secondary finding Our secondary result that superior behavioural performance on our visual target detection task was positively associated with lower falls risk concurs with the widely accepted cognitive profile of senior fallers. Evidence has consistently implicated cognitive impairment ? and more specifically, reduced executive  39 functioning ? in falls and falls risk (e.g., Hsu et al., 2012). In our study, task performance and falls risk were only significantly correlated during the more cognitively challenging condition (i.e., high load). This is consistent with previous reports that have demonstrated that fallers have impaired performance on complex, higher-level tasks requiring response inhibition and selective attention (e.g., Liu-Ambrose, Donaldson, et al., 2008; Lord & Fitzpatrick, 2001). In contrast, impaired performance on perceptually easy tasks ? akin to our low load condition ? is not characteristic of senior fallers (e.g., Woolley et al., 1997). Hence, our study lends further support to the notion that falls risk accompanies impaired executive cognitive functioning rather than a more generalized cognitive slowing.  For our electrophysiological results, we reported that an enhanced P300 mean amplitude to targets during the low load condition was associated with falls risk. Previous studies have found that the P300 modulates as a function of task difficulty, with a larger positive deflection during easy ? or low load ? tasks, relative to more difficult tasks (e.g., Kok, 2001; Polich, 1987). Predicated on the idea that the P300 has been proposed to index attentional resource allocation when memory updating is engaged (e.g., Polich, 1996), those at higher risk of falls may be more immersed in the primary task and less able to distribute attention. This idea certainly fits with the preexisting body of knowledge that has reported poor dual-task performance among fallers (e.g., Hsu et al., 2012). Most notably, these results apply to the low load condition, suggesting that falls risk may be associated with difficulty performing two tasks concurrently, even when the attentional demands of the primary task are low.  Additional issues An important question worth discussing is why we failed to observe modulations in attentional facilitation to peripheral probes as a function of load on the visual target detection task. Previous studies have reported increased visual-spatial attention ? as indexed by the P1 and N1 ERP components ? to peripheral stimuli  40 during low load conditions relative to high load (e.g., Handy et al., 2001). We highlight that our behavioural and electrophysiological data for the visual target detection task provide evidence that our paradigm manipulation was effective. Specifically, participants performed worse during the high load condition and P300 mean amplitude was significantly larger for low load blocks, consistent with our expectations (e.g., Dark, Johnston, Myles-Worsley, & Farah, 1985; Kok, 2001; Polich, 1987; Yantis & Johnston, 1990). Given that the capacity and distribution of attention resources are known to change with age (e.g., Craik & Byrd, 1982), future work regarding how attention is allocated to task-irrelevant peripheral probes in older adults is warranted.    We acknowledge the limitations of our study. First, we examined falls risk as a continuous variable, rather than categorizing participants as fallers versus non-fallers based on falls history. This is because memory for previous falls is subject to retrospective bias (Lachenbruch, Reinsch, MacRae, & Tobis, 1991). Our falls risk assessment using PPA scores was reliable, given that PPA was significantly correlated with number of falls over the past 12 months, balance confidence, and TUG time ? all of which discriminate fallers from non-fallers (e.g., Powell & Myers, 1995; Shumway-Cook, Brauer, & Woollacott, 2000). We also had a large range of falls risk within our sample, ranging from low risk to high risk. Second, the cross-sectional nature of our study does not allow us to infer a causal relationship between visual-spatial attention and falls risk. Thus, future intervention studies with large sample sizes are required to further elucidate the potential contribution of impaired visual-spatial attention to falls risk.  Conclusion In conclusion, our study is the first to our knowledge to provide evidence that reduced visual-spatial attention to task-irrelevant stimuli in the left visual field is associated with falls risk in older adults. It is well known that intact brain structure and function are essential for physical functioning. Surmounting evidence  41 is implicating cognitive function as an arbiter between the brain and mobility. Impaired mobility represents a major source of disability and loss of independence among older adults. With our rapidly aging population, identifying key contributors to falls risk is becoming an increasing health care priority. While the role of cognitive functions in falls risk is now being recognized more than ever, future research should focus on determining the exact nature of the relationship between visual-spatial attention, brain structure, and falls risk.     42 CHAPTER 3: Increased cognitive load leads to impaired mobility decisions in seniors at risk for falls Introduction Cognition is an important contributor to safe mobility through the environment. Although physical abilities (e.g., balance) undoubtedly factor into our capacity to be mobile, specific cognitive processes such as attention, planning, and decision-making collectively ensure our safety during mobility.   Within the multiple domains of cognition, executive functions ? or higher order cognitive processes ? are integral to safe mobility (e.g., Anstey et al., 2006; Persad et al., 1995). These include the ability to concentrate, attend selectively, and to plan and strategize. Selective attention to hazards in our environment is essential for safe mobility. Our ability to effectively plan and strategize contributes to decision-making about when and where to move. Importantly, a failure to make appropriate and timely decisions may result in unsafe mobility, such as falls.   With age, walking requires greater cognitive effort and a larger allocation of attentional resources (e.g., Lindenberger, Marsiske, & Baltes, 2000; Lovden, Schaefer, Pohlmeyer, & Lindenberger, 2008; Wollacott & Shumway-Cook, 2002; Yogev-Seligmann, Hausdorff, & Giladi, 2008). This likely results from reduced parietal cortex function, leading to a higher need for sensorimotor processing (e.g., Huxhold, Li, Schmiedek, & Lindenberger, 2006). Additionally, impaired prefrontal cortex function results in reduced employment of attentional resources for effective postural control (e.g., Huxhold et al., 2006).   Hence, not surprisingly, reduced executive functioning is a risk factor for falls (e.g., Tinetti et al., 1988). Falls, a clinical consequence of unsafe mobility and a significant health care problem, occur in  43 approximately 30% of community-dwelling seniors (e.g., Skelton & Todd, 2004; Tinetti et al., 1988). Falls are often attributed to impaired physical abilities but recent evidence now highlights the role of reduced executive functioning. Anstey and colleagues (2006) found that cognitive performance on ?Similarities?, a test of verbal reasoning, was inversely associated with falls rate. Also, performance on the Stroop Color-Word Test, a test of conflict resolution, predicts falls status beyond that explained by age and functional motor ability (Rapport, Hanks, Millis, & Deshpande, 1998). Both verbal reasoning and conflict resolution are dimensions of executive functioning (e.g., Lezak, 1995; Stuss & Alexander, 2000).   To better understand the interaction between executive functioning and falls risk, studies have relied on dual-task paradigms incorporating a motor task with a concurrent cognitive task. These studies consistently demonstrate that reduced gait speed during dual-task performance is associated with falls risk (e.g., Verghese et al., 2002) and that impairments in the ability to hold a conversation while walking are notable in senior fallers (e.g., Lundin-Olsson et al., 1997). However, studies to date have not examined the role of decision-making relevant to navigation and mobility during dual-task performance.   Previous studies are also limited by their use of laboratory-based dual-task paradigms. While dual-task paradigms such as reciting the alphabet or counting backwards while walking result in changes in overt motor outcomes (e.g., Verghese et al., 2002), they are not activities performed in the real world. Rather, in our modern and increasingly technology-based world, we are more likely to engage in conversations on a cell phone or listen to an iPod as we walk down the street.   To address these limitations, we examined the relationship between decision-making and falls risk in the context of a simulated real-world task using an immersive virtual reality environment (VRE). We compared seniors at-risk for falls with seniors not-at-risk on the ability to successfully cross a busy street under a  44 single-task and two different dual-task conditions. The primary objective was to determine if there were differences between the two groups in their ability to judge when it was safe to cross the street under dual-task conditions. The secondary objective was to determine if there were differences in gait speed between the two groups under single and dual-task conditions. Additionally, we included a computer-based dual-task to assess cognitive aspects of performing a dual-task independent of mobility.   Methods Participants Thirty-three community-dwelling seniors participated (16 female; Mean age = 73.12 years, SD = 4.46). Participants were recruited via advertisements and participant pools from previous studies within the lab at the University of Illinois. All interested participants were initially screened by phone. Inclusion criteria were: 1) community-dwelling, 2) age 65 years or older, 3) normal or corrected to normal vision, with visual acuity of 20/40 or better based on Snellen chart performance, 4) no diagnosis of a neurological or neuropsychological disorder, 5) not currently taking medication impeding balance, and 6) able to walk > 0.5 kilometers unaided.    Procedure There were two experimental sessions, each 1.5 hours. In Session 1, participants completed descriptive measurements, a computer dual-task paradigm, and falls risk assessment. In Session 2, participants performed the primary experimental task (i.e., VRE). Session order was counterbalanced between subjects. All participants provided written informed consent.      45 Measures Descriptive measures We measured age in years, standing height in centimetres, and mass in kilograms. We assessed global cognition using the MoCA, where scores > 26 indicate normal cognitive performance. Current level of physical activity was determined by the Physical Activities Scale for the Elderly (PASE) (Washburn, Smith, Jette, & al., 1993), where physical activities completed in the past 7-day period are reported. General mobility and balance were measured by the TUG (Podsiadlo & Richardson, 1991) and the SPPB.  Falls risk Physiological falls risk was assessed by the short form of the PPA (Lord et al., 2003). The PPA is a valid and reliable measure of falls risk in seniors, with 75% predictive accuracy for falls in older adults. Based on performance of five physiological domains (reaction time, contrast sensitivity, sway, proprioception, and knee extension strength), the PPA computes a falls risk score for each individual. We classified our participants as ?At-Risk? or ?Not-At-Risk? for future falls according to their PPA scores. Based on previous work demonstrating that a PPA cutoff score of 0.6 validly classifies seniors into separate falls risk categories, we decided a priori to divide our participants into two groups using this cutoff score (i.e., < 0.6 = ?Not-At Risk?; > 0.6 = ?At-Risk?) (Delbaere et al., 2010). While the TUG is another measure used to determine falls risk, we chose to divide our group based on PPA because it is currently the most valid way to assess falls risk (Lord et al., 2003).  Computer dual-task performance Cognitive dual-task ability was assessed using a computer-based paradigm. Participants viewed a computer display with either a single number or letter (single task) or a number and letter concurrently (dual task). Participants were required to respond by pressing a button with their left hand for letters and right  46 hand for numbers (index fingers corresponding to ?B? and ?2? and middle fingers corresponding to ?A? and ?3?, respectively). Reaction times and accuracy were recorded.   CAVE virtual environment Dual-task ability in simulated ?real-life? was assessed using the CAVE Virtual Environment (CAVE; Beckman Institute, Champaign-Urbana, Illinois http://isl.beckman.illinois.edu/Labs/CAVE/CAVE.htmlhttp://isl.beckman.illinois.edu/Labs/CAVE/CAVE.html). The CAVE consists of four viewing screens (one in the front, one on each side, and a floor), each measuring 303 cm wide x 273 cm high and a screen resolution of 1024 x 768 pixels. Participants stood approximately 149 cm away from the front screen, creating a viewing angle of 910 x 850. A custom designed program (Illinois Simulation Laboratory) provided the environment presentation, motion simulation, and data acquisition. Images were projected from a PC running on a 64-bit Windows Server (2003) and graphics were presented by an nVidia Quadro Plex 1000 Model 2. We monitored head movements (i.e., the number of times participants looked in either direction in preparation and while crossing the street) through an Ascension Flock of Birds 6DOF electromagnetic tracker, where head movements were defined as moving 100 in one direction to at least 100 in the opposite direction. Depth perception was created by wireless CrystalEyes liquid crystal shutter goggles, rapidly alternating the display to each eye, thus providing the ?virtual reality experience.?   Participants viewed a VRE simulating a two-way busy street, with cars approaching from both directions (Figure 3.1). Starting at a crosswalk, they had to cross two lanes of traffic, totaling eight meters in width. To cross the street, participants walked on a LifeGear Walkease manual treadmill, which was synced to the VRE. Cars moved at a constant speed of 54 km/hr, with consistent spacing of either 75 or 90 meters apart for each trial.  47  Figure 3.1. A screen-shot of the virtual reality display.  Note that the image was created from still captures of three separate images projected on the three walls of the CAVE (left, straight ahead, and right). In the virtual environment, the road appears as one straight line, perpendicular to the participant.    Participants were instructed to cross the street without getting hit by oncoming traffic. They were permitted to walk forwards only and could walk as fast as necessary without running. Trials began after passing through a gate, after which they had 90 seconds to complete the trial. There were eight practice trials. The actual experiment comprised sixty trials. Participants were permitted to take breaks between trials.   The experiment was a blocked design, with three experimental conditions: 1) No Distraction; 2) Music; and 3) Phone. Each block consisted of ten trials, and each of the three conditions was completed twice (i.e., twenty total trials per condition). Condition order was randomized. In the No Distraction condition (i.e., control condition), participants crossed the street without a secondary task. In the Music condition, participants listened to one of several pre-made playlists through earphones on an iPod while attempting to cross the street. In the Phone condition, participants crossed the street while conversing on the phone with Left Wall Right Wall Center Wall  48 a confederate via a hands-free headset. Experimenters had a set list of questions to engage the participants. An experimenter was present in the room to supervise the participant at all times.   Analysis Data was imported into SPSS (Version 16.0 for MAC). Descriptive measures and computer-based dual-task performance were analyzed using independent-samples t-tests to test for between-group differences. Primary CAVE measures were analyzed using repeated-measures ANOVAs, with condition as the within-subject factor, and group as the between-subject factor. We analyzed: (1) the number of successful trials completed, defined as crossing the street without being hit, (2) whether unsuccessful trials were the result of collisions versus running out of time, (3) number of collisions occurring in the first lane of traffic (assuming collisions in the first lane indicate reduced decision-making ability, or judgment, regarding when to initiate street-crossing), and (4) using only successful trials, length of time to cross the street. Body mass index (BMI; kg/m2) was included as a covariate for the analysis of length of time to cross the street, because stride length is an important contributor to gait speed (e.g., Callisaya, Blizzard, Schmidt, McGinley, & Srikanth, 2010; Kuo, Lin, Yu, Wu, & Kuo, 2009). For all analyses, the overall alpha level was set at p < 0.05.   Results Descriptive measures Table 3.1 provides results of the descriptive measures. Of these measures, both the TUG score and PPA score significantly differed between the groups, t(31) = 2.53, p = 0.02 and t(31) = 7.52, p < 0.001, respectively.      49 Table 3.1. Descriptive measures.  Measure1     ?Not-At-Risk? ?At-Risk? Effect sizes5 n = 17  n = 16  ?p2 Age, years     71.9 (4.1) 74.4 (4.6) 0.09  BMI, kg/m2     27.1 (4.2) 25.9 (3.2) 0.03  Education, No. (%)      No high school diploma   0 (0)  0 (0)      High school diploma    3 (17.6)  2 (12.5)      Some university certificate or diploma  6 (35.3)  3 (18.8)      University degree     2 (11.8)  4 (25.0)      Post-graduate    6 (35.3)  7 (43.8)  Female gender, No. (%)    8 (47.1)  8 (50.0)  Comorbidities, No.    1.53 (1.70) 1.75 (1.24) 0.01  PASE score     182.3 (82.0) 138.7 (61.3) 0.09  MoCA score2     24.9 (2.8) 24.1 (2.2) 0.03  Physical battery3    9.6 (1.4) 8.5 (1.7) 0.11  PPA score     -0.2 (0.5) 1.2 (0.6) 0.65**  TUG score, s4     9.8 (1.7) 11.4 (2.0) 0.17*  Abbreviations: PASE, Physical Activity Scale for the Elderly; MoCA, Montreal Cognitive Assessment. 1Unless otherwise indicated, data are expressed as mean (SD). Percentages have been rounded and may not total 100. 2Maximum is 30 points. 3Maximum is 12 points. 4Time recorded in seconds. 5Effect sizes calculated using ?p2. *p < 0.05 **p < 0.001    50 Computer dual-task performance ?At-Risk? seniors performed significantly worse in the dual-task compared to those ?Not-At-Risk? (Table 3.2). Specifically, ?At-Risk? individuals had both reduced accuracy, F(1,32) = 5.64, p = 0.02, ?p2 = 0.15 and slower reaction times, F(1,32) = 4.65, p = 0.04, ?p2 = 0.13 in the dual-task condition, relative to their ?Not-At-Risk? peers. In the single-task condition, there were no significant between-groups differences for either accuracy or reaction times (p = 0.11 and 0.93, respectively). Additionally, the number of successful crossing trials in the CAVE Phone condition was positively correlated with dual-task accuracy performance on the computer task, r = 0.48, p = 0.004. Furthermore, time to cross the street in the CAVE Phone condition was positively correlated with dual-task reaction time in the computer task, r = 0.54, p = 0.001.   Table 3.2. Task performance in the CAVE and on the computer-based dual-task as a function of condition.    ?Not-at-risk  ?At-risk?  CAVE1   Collision rate  Time out rate Collision rate Time out rate No distraction 20.88 (12.78) 0.59 (2.43) 29.06 (20.43) 0.94 (3.75)  Music  23.82 (13.29)  0.29 (1.21)  30.31 (18.12)  0.63 (2.50)  Phone  22.35 (15.12)  2.35 (6.15)  40.31 (20.29)  3.75 (11.18)   Computer-based dual-task2   Reaction time Accuracy Reaction time Accuracy  Single task 1099.81 (103.24) 85.60 (13.28) 1107.39 (313.77) 76.10 (19.75)  Dual task  1505.04 (121.97)  57.95 (19.95)  1585.66 (89.09)  40.33 (22.68)  1Data presented as mean percentage rates (SD). Rates calculated as number of unsuccessful trials divided by total number of trials x 100.  2Data presented as mean reaction time (s) and accuracy (% correct responses) (SD).  51 CAVE Trial success Regarding the number of times participants successfully crossed the street, there was a significant main effect of condition, F(2,62) = 6.65, p = 0.002, ?p2 = 0.18. Specifically, follow-up simple contrasts revealed that participants performed worse in the Phone condition compared to No Distraction, F(1,31) = 11.93, p = 0.002, ?p2 = 0.28. ?At-Risk? participants performed significantly worse overall than those ?Not-At-Risk?, as confirmed via a significant main effect of group, F(1,31) = 4.34, p = 0.05, ?p2 = 0.12. Additionally, there was a significant condition x group interaction, F(2,62) = 3.69, p = 0.01, ?p2 = 0.11. Follow-up analyses revealed that ?At-Risk? participants successfully crossed the street significantly fewer times in the Phone condition relative to those ?Not-At-Risk?, F(1,32) = 7.86, p = 0.009, ?p2 = 0.20, but no significant between-groups differences were found for the No Distraction (p = 0.17) and Music (p = 0.25) conditions (Figure 3.2.1).      52 Figure 3.2.1. Street crossing performance as a function of condition (No distraction, Music, and Phone) and falls-risk group.  Error bars represent standard errors of the mean. Mean number of trials successfully completed. Those ?At-Risk? for falls successfully crossed the street significantly fewer times than those ?Not-At-Risk? in the Phone condition.     For unsuccessful trials, we examined rates of collisions versus time-outs (Table 3.2). Collisions were defined as trials where participants were struck by an oncoming car during crossing. Time-outs were defined as trials where participants were unable to cross the street within the allotted 90-second time, and were not struck. For collisions, there was a significant main effect of condition, F(2,62) = 3.26, p = 0.05, ?p2 = 0.10. Follow-up simple contrasts reveal that overall, participants experienced more collisions in the Phone condition relative to No Distraction, F(1,31) = 7.24, p = 0.11, ?p2 = 0.19. There was also a significant main effect of group, F(1,31) = 4.57, p = 0.04, ?p2 = 0.13, where ?At-Risk? participants experienced more collisions than those ?Not-At-Risk.? Lastly, there was a marginally significant condition x group interaction, F(2,62) = 12.64, p = 0.06, ?p2 = 0.09. Follow-up analyses show that ?At-Risk? participants experienced more collisions in the Phone condition, relative to age-matched controls, F(1,32) = 8.39, p = 0.007, ?p2 = 0.21. There were no significant between-groups differences for the No Distraction (p = 0.18) and Music (p =  53 0.25) conditions. For time-outs, there was a significant main effect of condition, F(2,62) = 4.58, p = 0.01, ?p2 = 0.13, with more time-outs observed for the Phone condition relative to No Distraction, F(1,31) = 4.89, p = 0.04, ?p2 = 0.14. There were no significant between-group differences, p > 0.05 for number of time-outs.   Street-crossing time For the length of time taken to cross the street, there was a significant main effect of group, F(1,30) = 4.68, p = 0.04, ?p2 = 0.14, indicating that overall, those ?At-Risk? for falls were significantly slower to cross the street. More specifically, however, ?At-Risk? participants crossed the street significantly slower in the Phone condition compared to those ?Not-At-Risk? (Figure 3.2.2), as indicated by the significant condition x group interaction, F(2,60) = 3.45, p = 0.04, ?p2 = 0.10. Follow-up analyses revealed significant between-group differences in the Phone condition, F(1,32) = 9.00, p = 0.005, ?p2 = 0.23. There were no significant between-groups differences for the No Distraction (p = 0.12) and Music (p = 0.14) conditions.      54 Figure 3.2.2. Street crossing performance as a function of condition (No distraction, Music, and Phone) and falls-risk group.  Error bars represent standard errors of the mean. Mean length of time taken to cross the street on successful trials. Those ?At-Risk? for falls crossed the street significantly slower than those ?Not-At-Risk? in the Phone condition.   Discussion To examine the relationship between mobility judgments and falls-risk in the real-world, participants ?At-Risk? for falls and those ?Not-At-Risk? crossed a busy street in a VRE while completing a secondary task. We report two key results. First, seniors ?At-Risk? were less successful at street-crossing while conversing on a phone. Reduced success rate was secondary to greater number of collisions in the first lane of traffic. Second, those ?At-Risk? crossed the street significantly slower compared with those ?Not-At-Risk? in the Phone condition. Given these results, several points of discussion follow.   First, our results are consistent with previous findings that dual-task performance leads to reduced gait speed in senior fallers (e.g., Faulkner et al., 2007; Lundin-Olsson, Nyberg, & Gustafson, 1998; Verghese et al., 2002; Yogev-Seligmann et al., 2008). In tasks that require performance of a physical task (e.g., walking)  55 and a concurrent cognitive task (e.g., talking), young adults tend to prioritize gait and compromise cognitive performance (e.g., Bloem, Valkenburg, Slabbekoorn, & Willemsen, 2001; Yogev-Seligmann et al., 2008). However, Parkinson?s patients and senior fallers are susceptible to increased gait variability and reduced ability to prioritize gait performance under dual-task conditions (e.g., Beauchet et al., 2007; Bloem, Valkenburg, Slabbekoorn, & van Dijk, 2001; Chapman & Hollands, 2007; Yogev-Seligmann et al., 2008). It has therefore been suggested that reduced executive functioning leads to impaired divided attention and ineffective use of available resources, resulting in reduced gait speed (e.g., Yogev-Seligmann et al., 2008).   More notably, our study is the first to show that individuals ?At-Risk? for falls may have reduced ability to plan and decide on their mobility through the physical environment when cognitively loaded. Reduced abilities to plan and judge under dual-task conditions among seniors ?At-Risk? for falls are likely to be secondary to reductions in cognitive capacity evident in this population (e.g., Rapport et al., 1998; Springer et al., 2006). Reduced judgment resulting from increased cognitive load may result in two forms of behaviour: 1. Conservative behaviour, or 2. Risky behaviour. A previous study used the same VRE paradigm to examine the effects of dual-task performance on pedestrian behaviour among college-aged adults (Neider, McCarley, Crowell, Kaczmarski, & Kramer, 2010). While talking on a cell phone, young adults were more cautious crossing the street, such that participants took longer to both initiate and complete street-crossing. Young adults also timed-out more during the phone condition. Hence, dual-task demands resulted in more cautious behaviour among young adults. In contrast, seniors ?At-Risk? for falls engaged in risk-taking behaviour, jeopardizing their safety. Specifically, they had significantly more collisions in the first lane than those ?Not-At-Risk?, suggesting that seniors at risk for falls were less able to appropriately judge when to initiate street crossing. Interestingly, in a previous study senior fallers cited their own risk-taking behaviours as the most common cause of falling, rather than their health or environmental factors (e.g., Hornbrook, Wingfield, Stevens, Hollis, & Greenlick, 1991). Hence, an inability  56 to judge one?s own neuromuscular constraints to plan successful movements, as limited by executive functioning impairments, may be linked to falls-risk (e.g., Liu-Ambrose, Ahamed, Graf, Feldman, & Robinovitch, 2008).   Our results also suggest that dual-task demands per se, may not be detrimental to safe mobility. Rather, the nature of the concurrent task demands (i.e., a ?passive? task, such as listening to music versus an ?active? task, such as talking on the phone) is important. This factor may have contributed to the equivocal results to date on the association between dual-task performance and falls risk. For example, Verghese and colleagues (2002) found that reciting the alphabet while walking significantly reduced gait speed in fallers. In contrast, Bootsma-van der Wiel et al. (2003) asked participants to count backwards while walking, and found that dual-task performance is not a significant predictor of falls. Hence, future studies are needed to better ascertain the modulating effect of cognitive load on the relationship between dual-tasking ability and falls risk.   Finally, while there are certainly differences in mobility between those ?At-Risk? and those ?Not-At-Risk? for falls, our results were not merely due to physical differences between the two groups. First, there were no group differences in the SPPB and current physical activity level. Second, gait speed was not significantly different between the two groups in both the Music and No Distraction conditions. Third, results from the computer-based dual-task support the notion that our results are likely due to between-groups differences in cognitive control, as results from the computer task and the CAVE data positively correlate. Together, our findings suggest that differences between our falls-risk groups can be attributed, at least in part, to cognitive abilities.    57 We recognize the limitations of our study. First, due to recruitment issues, we were unable to directly assess seniors with a history of falls (i.e., ?fallers?) versus seniors without a history of falls (i.e., ?non-fallers?). Instead, our groups were defined using a validated PPA cutoff score to identify those ?At-Risk? versus those ?Not-At-Risk? for future falls (Lord et al., 2003). Second, our ability to make conclusions regarding the Music condition is limited by our lack of behavioural measures of the secondary task. Specifically, it is possible that participants were simply ?tuning-out? the music, performing only the walking task without a secondary cognitive task. However, listening does not appear to impair secondary task performance (McCarley et al., 2004). Therefore, our results are more likely due to actual differences between listening and talking, rather than mere task engagement. Another limitation was that we did not compensate for the fact that our viewing distance requires a small accommodative response in order to yield a well-focused retinal image. This may be addressed in future studies by adjusting the LCD stereo goggles. A fourth limitation is that our virtual environment did not replicate any traffic sounds. Future studies may examine how inclusion of road noise may affect street-crossing performance by providing auditory cues. Lastly, our task was designed to be more difficult than crossing the street in real life. Indeed, street-crossing performance in all conditions was markedly lower than what we would expect in real life. However, this was necessary in order to ensure that there would be performance variability between participants. While we recognize that this is certainly a limitation of our current study, we highlight that our study has increased ecological validity, compared to strictly laboratory-based tasks, and that our results here represent the trade-off between internal and external validity.   To conclude, our study suggests that critical mobility decisions, such as when to cross a busy street, may be impaired in those even at moderate risk for falls (i.e., mean PPA score of 1.2). These impaired judgments result from increased cognitive load. We highlight the value of increasingly ecologically-valid  58 paradigms designed to test ?real-life? situations. Given the complex relationship between cognitive and physical abilities, it is important to understand how they may interact in the context of the real world.      59 CHAPTER 4: Mind-wandering and falls risk in older adults: Evidence for failures in compensatory cognition Introduction The natural tendency for our thoughts to drift off task ? known as mind-wandering ? has become an increasingly popular topic of research in neuroscience (e.g., Smallwood, in press). Although it is a ubiquitous phenomenon, with up to 50% of our waking time spent creating and maintaining an inner dialogue secondary to current behavioural goals (Smallwood, in press), variations in mind-wandering are associated with neurocognitive pathologies such as dysphoria (Smallwood, O'Connor, Sudbery, & Obonsawin, 2007) and attention deficit hyperactivity disorder (ADHD) (e.g., Shaw & Giambra, 1993). Furthermore, previous reports suggest that mind-wandering frequency is modulated as a function of age ? with older adults spending significantly less time engaging in task-unrelated thoughts (e.g., Giambra, 1989; Jackson & Balota, 2012). Given what we now understand about how mind-wandering impacts neurocognitive functioning, the aim of our study was to establish whether alterations in mind-wandering may be also contributing to one of the primary health risks of aging ?? namely, falling.  Evidence that mind-wandering may be a heretofore unrecognized risk factor of falling in older adults stems from three core findings regarding the transient effects of mind-wandering on neurocognitive function. First, mind-wandering directly alters how we perceive, analyze and respond to the external environment. This statement is substantiated by neuroimaging evidence using ERPs. In a study examining the effects of mind-wandering on sensory gain control, Kam and colleagues (2011) found that during off-task periods, sensory processing was attenuated. Specifically, sensory-evoked responses to stimuli presented in both visual and auditory modalities were reduced during periods of mind-wandering, relative to ?on-task? (or non-mind-wandering) states. In addition, Smallwood et al. (2008) reported decreased cognitive processing of  60 visual stimuli during mind-wandering, as measured by the P300 ERP component. Together, these studies converge on the idea that there is a systematic reduction in the depth of stimulus processing at both the sensory/perceptual and cognitive levels. Likewise, behavioral control shifts to a more automatic state (e.g., Carriere, Cheyne, & Smilek, 2008; Cheyne, Carriere, & Smilek, 2006; Reichle, Reineberg, & Schooler, 2010; Smallwood et al., 2008), leading to speeded responses and higher error rates on task performance during mind-wandering compared to on-task periods (Franklin, Smallwood, & Schooler, 2011; Smallwood et al., 2004). In short, mind-wandering may impact a host of neurocognitive functions essential for safe navigation and mobility. This predicts that the more an older individual mind-wanders, and/or the greater the extent of these effects on the individual?s neurocognitive function, the more susceptible that individual may be to falling.  Second, the propensity to mind-wander has been directly linked to executive cognitive functioning and its functional capacity. In particular, the executive failure hypothesis of mind-wandering posits that higher-level executive control is required to sustain external attention, and at the same time, ignore internal and perceptual distractions (Smallwood, in press). Evidence suggests that seniors with a history of falls have impaired executive functioning, including poorer performance on tasks involving response inhibition and selective attention (e.g., Liu-Ambrose, Donaldson, et al., 2008; Lord & Fitzpatrick, 2001; McGough et al., 2011; Springer et al., 2006). Thus, an inability to actively control attention to align with current behavioural goals, while simultaneously inhibiting task-unrelated thoughts, may represent a major obstacle for fallers to safely navigate their environment. Furthermore, mind-wandering results in impaired performance on the primary task (e.g., Grodsky & Giambra, 1990-1991). Such performance decrements are akin to the dual-task costs often observed in healthy adults, which are further exacerbated in senior fallers (Lundin-Olsson et al., 1997; Nagamatsu, Voss, et al., 2011; Verghese et al., 2002; for a review, see Hsu et al., 2012).  61 Importantly, dual-task impairments are indicative of reduced general processing capacity (e.g., Kahneman, 1973), which may in turn be linked to a greater frequency of mind-wandering (e.g., Kane et al., 2007).  Lastly, mind-wandering, or regular oscillations in the depth of our neurocognitive engagement with the external environment, is normative to healthy human brain function (Schooler et al., 2011; Smallwood, in press; Smallwood & Schooler, 2006). However, increased frequency of mind-wandering has now been tied to several clinical and sub-clinical neurocognitive pathologies. These include clinical signs of depression, such as dysphoria (Smallwood et al., 2007; see also Killingsworth & Gilbert, 2010), ADHD (e.g., Shaw & Giambra, 1993), impulsivity (e.g., Helton, 2009), and schizophrenia (e.g., Elua, Laws, & Kvavilashvili, 2012). On the other hand, seniors show a decreased propensity for mind-wandering relative to young adults (e.g., Giambra, 1989, 1993; Jackson & Balota, 2012), which is thought to reflect task-engagement (Jackson & Balota, 2012). It has also been suggested that mind-wandering frequency is directly associated with processing demands of the current task (Mason et al., 2007); thus, reduced mind-wandering in older adults may reflect a strategy aimed at compensating for reduced cognitive capacity as a function of age. Briefly put, because cognitive capacity is known to decline with age (Craik & Byrd, 1982), older adults must dedicate a greater proportion of their available resources to the task at hand ? thus leaving fewer resources to allocate to a secondary task, such as task-unrelated thoughts. Overall, such results indicate that not only are altered patterns of mind-wandering associated with neurocognitive pathology, but that any increased propensity for mind-wandering in senior fallers would be in direct opposition to what is normative for their age group.  In light of the above considerations, our study was designed to examine two different ways in which mind-wandering could be associated with falls risk in seniors: 1) The frequency rate or relative amount of time spend mind-wandering; and 2) The qualitative/quantitative way in which mind-wandering impacts  62 neurocognitive function. Based on the predictions outlined above, our specific goal was to determine whether falling in seniors may be associated with an increased tendency to mind-wander relative to seniors with no falling history, significantly greater impacts of mind-wandering on stimulus processing, or both. Accordingly, participants completed a sustained visual target-detection task (sustained attention to response task, SART) while ERPs were recorded. We determined the rate or amount of mind-wandering for each participant via subjective reports of participants current attentional state (on-task versus mind-wandering), collected at regular but unpredictable intervals (see below). To determine the impact of mind-wandering on neurocognitive processing, we examined the ERPs elicited by targets as a function of whether they occurred during an ?on-task? versus ?mind-wandering? state.  The SART is a widely-used paradigm in mind-wandering studies, where participants are required to respond to frequently presented targets, while inhibiting responses to infrequent targets (e.g., Kam et al., 2011; Smallwood et al., 2008). The SART is thought to measure sustained attention ? which is influenced by the moment-to-moment efficacy of attentional control mechanisms (Manly, Robertson, Galloway, & Hawkins, 1999). Due to the repetitive and mundane nature of the task, the SART is well-suited to induce periods of mind-wandering over a sustained testing period. Furthermore, SART performance itself can be a measure of attentional state, in that errors of commission are more likely to occur when one is off-task and thus unable to inhibit the prepotent response (Smallwood et al., 2004). Indeed, poor performance on the SART has been attributed to absent-mindedness (Manly et al., 1999; Robertson, Manly, Andrade, Baddeley, & Yiend, 1997). Previous ERP studies using the SART ? and more generally, go-no-go paradigms ? have focused on the N200 and P300 ERP components. First, the anterior N200 ? a negative deflection in the ERP waveform occurring approximately 200 msec post-stimulus at frontal electrode sites ? has been implicated in attentional control (e.g., Folstein & Van Petten, 2008), and is larger in amplitude for infrequent ?no-go? trials, in comparison to frequent ?go? trials. Second, the P300 is characterized by a large  63 positive deflection in the waveform approximately 300 msec following stimulus presentation at central-parietal electrode sites. The P300 indexes higher-level cognitive evaluation of stimuli (e.g., Donchin & Coles, 1988) and attenuates as a function of mind-wandering (Smallwood et al., 2008).  With respect to identifying the attentional state of our participants during task performance, we relied on ?experience sampling? (Schooler et al., 2011). Considered to be a ?direct? measure of mind-wandering, experience sampling relies on the fact that if prompted, we can reliably report on the content of our thoughts at any given moment, and further, determine whether they center on the on-going task being performed (referred to as an ?on-task? state), or alternatively, whether they have drifted off to unrelated issues (referred to as an ?off-task? or ?mind-wandering? state) (for a review, see Gruberger, Ben-Simon, Levkovitz, Zangen, & Hendler, 2011). Although the act of reporting on one?s attentional state interferes with the content of consciousness itself (e.g., Filler & Giambra, 1973), by using the report to categorize a participant?s attentional state in the 10-15 seconds immediately prior to the report, the methodology has been used to demonstrate reliable and replicable differences in neurocognitive functioning between ?on-task? and ?off-task? states (e.g., Christoff, Gordon, Smallwood, Smith, & Schooler, 2009; Franklin et al., 2011; Kam et al., 2011; Kirschner, Kam, Handy, & Ward, 2012; McKiernan, D'Angelo, Kaufman, & Binder, 2006; Smallwood et al., 2008; Smallwood et al., 2004; Stawarczyk, Majerus, Maj, Van der Linden, & D'Argembeau, 2011). As such, in adopting this methodology here, our approach to defining attentional states aligned with widely-accepted norms in the field of mind-wandering research.  Here we outline our predictions for our current study. For our behavioural results, we hypothesized that falls would be positively associated with frequency of mind-wandering, given that both have underlying relationships with reduced attentional control. Furthermore, while we did not anticipate reaction times to be associated with falls (e.g., Nagamatsu et al., 2009), we did hypothesize that falls would be associated with  64 task accuracy ? or errors of commission ? based on the idea that older adults with a history of falls have poorer response inhibition (Liu-Ambrose, Donaldson, et al., 2008; Liu-Ambrose, Katarynych, Ashe, Nagamatsu, & Hsu, 2009; Lord & Fitzpatrick, 2001; McGough et al., 2011). Regarding the neural signatures of cognitive processing, we examined mean amplitudes of the P200, N200, and P300 ERP components. Two main effects were expected: 1) That we would observe modulations in the amplitude of the three key components as a function of attentional state (e.g., Smallwood et al., 2008); and 2) That these modulations would be associated with history of falls over the past 12 months.  Methods Participants Eighteen community-dwelling senior women participated in this study. Participants were recruited from a database of people who had participated in previous studies in our laboratory, and who agreed to be contacted to participate in future research. They had previously been recruited through advertisements in the community and local media, and were screened for neurodegenerative disease and depression. Because our previous studies focused on the effects of physical activity on cognitive functions in older adults, and exercise is known to differentially impact cognition between the sexes (e.g., Baker et al., 2010), our sample only included women. Participants were aged 66 to 81 years (mean age 71.89 years, SD = 4.17). All participants were cognitively intact, as indicated by MMSE scores above 24 (Folstein, Folstein, & McHugh, 1975). Fifteen participants were right handed. All participants had normal or corrected-to-normal vision. Participants provided written informed consent and the reported research was approved by the Clinical Research Ethics Board at the University of British Columbia.      65 Falls We recorded the number of falls experienced by each individual over a twelve-month period prior to study participation. Falls were recorded using monthly falls calendars, where participants were required to mark on a calendar each day whether a fall occurred that day or not. Falls calendars were mailed into our study centre at the end of each month. Any falls that did occur that month were followed up via telephone interview to ascertain the circumstances regarding the fall and whether any injuries were sustained. Importantly, falls calendars are considered a valid and reliable method to track falls, rather than relying on retrospective reports which are subject to memory bias (Hannan et al., 2010; Lamb, Jorstad-Stein, Hauer, & Becker, 2005).  Stimuli and procedure Participants completed the SART used in previous mind-wandering studies (Kam et al., 2011; Smallwood et al., 2008). Trial sequence and timing are provided in Figure 4.1. Stimuli were presented on a computer monitor with a 24 cm diameter viewing screen, placed 110 cm from the participant. During each trial, participants were presented with a single visual stimulus. Participants viewed frequently presented visual targets at fixation (i.e., a number from 0-9, presented in a random sequence between trials) and were asked to respond as quickly and accurately as possible to their appearance with a manual button press using their right thumb. Infrequent visual targets (?non-targets?; i.e., the letter ?X?) were presented either once or twice per experimental block (i.e., half of blocks had one non-target, while the other half had two non-targets), where participants were required to withhold their response. In blocks with two non-targets, the ?X??s were separated by at least ten events. In addition, task-irrelevant visual probes (i.e., a box with  66 horizontal lines) were presented in the upper and lower visual periphery following each target 1. Participants were instructed to ignore these probes, as they were not directly relevant for their task.   Figure 4.1. Trial sequence and timing of the SART paradigm.                                                           1 The probes were included in the paradigm to replicate the task used by Kam et al. (2011). While we did examine amplitudes of the P1 ERP component time-locked to the probes, there was excessive residual noise in the waveforms, rendering them uninterpretable. Indeed, previous work with the P1 and older adults have shown that ERP morphology is impacted by age, with less reliable signals present in seniors (Curran, Hills, Patterson, & Strauss, 2001; Nagamatsu, Carolan, Liu-Ambrose, & Handy, 2011). Thus analyses related to the probes have been excluded from this manuscript. ??????????????? ?? ??????? ?? ??? ?? ??? ?? ? 67 The end of each block was signaled with a blue coloured screen, upon which participants were asked to verbally self-report their current attentional state at the time of the blue screen appearing ? whether they were on-task or mind-wandering. The experimenter noted their attentional state for each block, and these reports were later used to categorize the behavioural and electrophysiological data to compare the two attentional states. ?On-task? was defined as thoughts that were exclusively focused on the central experimental task. ?Mind-wandering?, in contrast, was defined as engagement in thoughts external to the experimental task. This distinction was explained to participants at the beginning of the experiment. To reduce demand characteristics, participants were also reassured that mind-wandering during the experiment was natural, and that there would be no negative consequences of doing so. There were 40 experimental blocks altogether. Each block was randomly selected to be 30-90 seconds long (approximately 15-45 trials), with two goals in mind: 1) To minimize predictability of block completion; and 2) To maximize variability in attentional state at the time of block completion. Breaks were permitted between blocks, as required by participants. We instructed participants to keep their eyes on the central fixation point for the duration of the experiment.  Electrophysiological recording and analysis EEGs were recorded from 64 active electrodes (Bio-Semi Active 2 system) evenly distributed over the head. All EEG activity was recorded relative to two scalp electrodes located over medial-frontal cortex (CMS/DRL), using a second order low pass filter of 0.05 Hz, with a gain of 0.5 and digitized on-line at a sampling rate of 256 samples-per-second. To ensure proper eye fixation and allow for the correction and/or removal of events associated with eye movement artifacts, vertical and horizontal EOGs were also recorded, the vertical EOG from an electrode inferior to the right eye, and the horizontal EOG from an electrode on the right outer canthus. Off-line computerized artifact rejection was used to eliminate epochs during which detectable eye movements (>10), blinks, or muscle potentials; custom threshold limits were  68 set on an individual-basis to reject the appropriate trials. ERPs for each condition of interest were averaged into 3000 msec epochs, beginning 1500 msec before stimulus onset. Next, all ERPs were algebraically re-referenced to the average of the left and right mastoid signals, and filtered with a low-pass Gaussian filter (25.6 Hz half-amplitude cutoff) to eliminate any residual high-frequency artifacts in the waveforms.    Electrophysiological analysis was performed using ERPSS (http://sdepl.ucsd.edu/erpss/doc/index.html). Statistical quantification of ERP data was based on mean amplitude measures relative to a -200 to 0 pre-stimulus baseline. The waveforms for each condition of interest (i.e., blocks where participants reported being on-task versus mind-wandering) were based on averaging together the EEG epochs for the six targets preceding each self-report, representing approximately 12 seconds of recording. Current evidence suggests that this time window accurately captures the length of time that one is in any given attentional state (e.g., Christoff et al., 2009; Kam et al., 2011; Sonuga-Barke & Castellanos, 2007). Importantly, our paradigm was designed to ensure that there were not any non-targets (i.e., ?no-go?) trials within the last 12 seconds of each block. Thus, all EEG activity was time-locked only to targets (i.e., ?go?) trials.  Because ERP component time windows and topography can vary as a function of age and task complexity, our parameters were chosen based on visualization of our data. Specifically, for each ERP component, time windows were chosen based on peak amplitude for each identified component, as visualized in the ERP waveforms. Electrode sites for analysis were based on peak activity shown on topographic maps of all electrodes on the scalp. We analyzed our ERP data using repeated measures mixed-model ANOVAs within ERPSS. For each component, our models included attentional state (on-task, mind-wandering) and electrode as within-subject factors. Our primary comparison of interest was the main effect of attentional state ? that is, whether the component amplitude was significantly modulated by being on-task versus mind-wandering. To ascertain the relationship between mind-wandering and falls, we calculated Pearson  69 correlations for both our behavioural measures (reaction times, accuracy, and frequency of mind-wandering) and electrophysiological measures (mean amplitudes of each ERP component) with number of falls over the past 12 months. Correlations were performed using SPSS (Version 20, MAC).  Physical falls risk factors To examine the relationship between falls and mind-wandering above and beyond physical falls risk factors, we included physiological measures in our study. First, we assessed gait speed, where we recorded the time required to walk four metres. Second, we assessed general cardiovascular capacity using the Six Minute Walk Test (6-MWT) (Enright, 2003), where the total distance walked (metres) in six minutes is measured. Third, we administered the TUG (Podsiadlo & Richardson, 1991), which requires participants to rise from a seated position, walk three metres, return to the chair, and sit down. The average time to complete each of two trials is recorded. Last we had participants complete the PPA (Lord et al., 2003). This is a valid and reliable measure of physiological falls risk based on a composite score from five distinct measures.  Results Falls Descriptive measures for all participants are reported in Table 4.1. During the 12-month period that falls were reported, 13 out of 18 participants experienced one or more falls. The number of falls reported by each participant varied from zero to four falls, with an average of 1.17 falls per participant (SD = 1.10 falls).        70 Table 4.1. Descriptive characteristics.  Variable1 All participants  (n = 18) EEG subset  (n = 15)  Mean (SD)  Mean (SD)  Age, years 71.89 (4.17) 70.53 (2.80) MMSE2 27.78 (2.10) 27.87 (2.23) Falls (past 12 months), number 1.17 (1.10) 1.33 (1.11) Gait Speed, s 3.40 (0.72) 3.28 (0.74) 6-MWT, m3 520.29 (91.64)  526.79 (96.68) TUG, s 7.00 (1.78) 6.88 (1.88) PPA4 0.38 (0.65) 0.29 (0.68) 1Unless otherwise indicated, data are expressed as mean (SD).  2Maximum was 30 points. 3One participant in the EEG subset did not complete the Six Minute Walk Test for safety reasons. 4Expressed as a z score indicating relative falls risk.  Behaviour For subjective reports, our participants reported mind-wandering during an average of 12.39 (SD = 6.01) out of 40 ? or 31% ? of blocks. Participants were able to successfully withhold their response from non-targets 68.89% (SD = 14.11%) of the time. For reaction times, participants responded slightly faster during periods of mind-wandering (mean = 418.16 msec, SD = 83.77 msec) compared to being on-task (mean = 430.46 msec, SD = 96.37 msec), although the difference was not significant, t(16) = 0.67, p = 0.51.  Before computing the correlations between our behavioural variables, we tested our data for skewness and the presence of outliers. We determined that falls history and our behavioural measures fit within the  71 acceptable range for skewness (highest skew value = 1.56 for reaction times). For outliers, there were no participants that were + 3 standard deviations from the mean. While there were participants + 2 standard deviations from the mean, they were not systematically outliers for multiple variables (i.e., one participant was an outlier for reaction times, another was an outlier for accuracy). Thus, we did not have adequate justification for removing these participants from our dataset. Our correlational data are presented in Figure 4.2. For our correlational analysis, the number of falls experienced over our 12-month observation period was associated with frequency of mind-wandering. This was established via a significant positive correlation between number of falls and number of subjective mind-wandering reports, r(18) = 0.47, p < 0.05. In addition, there was a significant negative correlation between reaction time to targets and number of falls, r(18) = -0.48, p = 0.04, such that a larger number of falls was associated with significantly faster reaction times on the SART task. There was also a significant negative correlation between accuracy (correct rejection of non-targets) and falls, r(18) = -0.63, p = 0.005. Specifically, more falls over the past year were associated with lower accuracy scores on our sustained attention task.     72 Figure 4.2. Scatterplots of significant correlations.  Number of falls over the 12 month assessment period as a function of (from top to bottom): percentage of blocks that participants self-reported mind-wandering, average reaction time on the SART task, and accuracy on the SART task.     ??????????????????? ? ? ? ? ?????????????????????????????????????? ? ? ? ? ????????????????????????????????????????? ? ? ? ? ???????????????? 73 Electrophysiology Three participants were excluded from our electrophysiological analyses due to excessive noise in their data at the individual level. Therefore, our sample for the ERP data consisted of 15 participants. ERP waveforms and topographies are presented in Figures 4.3 and 4.4. First, we examined the mean amplitude of the P200 component at 150-250 msec post-stimulus, centered on the peak amplitude. We used midline electrode sites, FZ, FCZ, CZ, and CPZ, and time-locked the signal to targets presented at fixation. There was no difference in mean P200 amplitude between attentional states, as indicated via a non-significant main effect, F(1,14) = 0.00, p = 0.99. Second, we examined the mean amplitude of the N200 component during a time window of 280-370 msec post-stimulus at electrode site AFZ. There was a significant main effect of attentional state on N200 amplitude, F(1,14) = 4.72, p < 0.05. Specifically, mean N200 amplitude was larger during periods of mind-wandering relative to when participants were on-task. The third component we examined was the P300, using a time window of 410-510 msec post-stimulus, centered on the peak amplitude, at central-parietal electrode sites, CZ, CPZ, C1, C2, CP1, and CP2, and time locked to targets presented at fixation. There were no significant modulations of the P300 for attentional state, F(1,14) = 0.34, p = 0.57.     74 Figure 4.3. Grand averaged waveforms for our ERP data (n = 15).  We examined the P200, N200, and P300 components. Significant modulation of the N200 was found as a function of attentional state (highlighted in grey).  ? ? ?? ? ???? ??? ???? ????? ??? ??? ??? ???? ??? ??? ?????? ?? ? ?? ?? ?????? ????? ????? 75 Figure 4.4. Topographies for ERP components as a function of component and attentional state.  ???? ? ????? ????????????? 76 There were no significant relationships between our electrophysiological measures and frequency of falls. To the point, our correlational analysis revealed that mean amplitudes of each of our three ERP components were not related to the number of blocks participants reported mind-wandering. Specifically, our analysis included mean amplitudes for all electrodes used individually for each component (all p > 0.16), average of the electrodes used for each component (all p > 0.44), and the difference in amplitude between on-task and mind-wandering blocks for the averaged electrodes (all p?s > 0.32).  Physical falls risk factors Descriptive measures of physical falls risk factors are presented in Table 4.1. We examined the correlations between our physical measures of falls risk factors and both number of falls and frequency of mind-wandering. None of our physical measures were significantly correlated with either number of falls (all p?s > 0.08) or frequency of mind-wandering (all p?s > 0.41).  Discussion Our study was designed to examine whether falls are associated with an increased tendency to engage in task unrelated thoughts, and whether this association might co-occur with differences in the neurocognitive processing of external stimuli during periods of mind-wandering. Towards answering this question, our results revealed a key relationship between history of falls and frequency of mind-wandering, where more falls were associated with a higher proportion of time spent off-task. Moreover, falls were significantly correlated with poorer behavioural performance on the sustained attention task. That is, faster reaction times and reduced accuracy were observed in those who experienced falls over the past year. In contrast, we did not find differences in the impact of mind-wandering on neurocognitive processing as a function of falls, as measured via ERPs. Taken together, our findings align with prevailing theories on mind-wandering and provide insight into the cognitive issues that contribute to falls in seniors.  77  Our finding that mind-wandering frequency is associated with number of falls is consistent with the impairments in executive cognitive functioning that are commonly observed in fallers. The relationship between proportion of time spent engaging in thoughts peripheral to the task and falls may be mediated by poor attentional control (e.g., Smallwood, in press). This is substantiated by recent findings that people who score poorly on measures of attentional control tend to engage more in task-unrelated thoughts (e.g., Kane et al., 2007; McVay & Kane, 2009). Based on our results, one key question that is raised is why mind-wandering, a common daily occurrence, might be associated with such a detrimental outcome in seniors. We know that aging is characterized by a host of cognitive changes, including declines in memory (e.g., Moscovitch, 1992), visual-spatial attention (e.g., Greenwood, Parasuraman, & Alexander, 1997), and general processing capacity (e.g., Wright, 1981). As such, neuroimaging evidence points towards neural activation aimed at compensating for these deficits. Reduced lateralization illustrated by the HAROLD (Hemispheric asymmetry reduction in older adults) model, for example, allows the recruitment of secondary regions in the brain to provide support to areas which may no longer be sufficient to process the required information (e.g., Cabeza, 2002). Therefore, engaging in less frequent mind-wandering (e.g., Giambra, 1989; Jackson & Balota, 2012) may represent a compensatory strategy in seniors. We speculate that this strategy may not be available to all seniors, resulting in more mind-wandering, less available processing capacity, and as a consequence, increased likelihood of falls.  That we found impaired behavioural performance on the sustained attention task in fallers is not surprising given that higher mind-wandering frequency and reduced accuracy come hand-in-hand (e.g., Giambra, 1995; Grodsky & Giambra, 1990-1991). Interestingly, falls were also associated with faster reaction times, providing evidence for a speed-accuracy trade-off among fallers and suggesting that they may be operating on ?pilot mode? during task performance. These results have important implications for seniors; Poor  78 behavioural performance in the context of every day life can mean failing to notice hazards and obstacles in the environment pertinent to safe navigation. Indeed, up to 30% of community-dwelling older adults experience one or more falls per year (e.g., Tinetti et al., 1988), representing a major health-care concern for our aging population. Increasing meta-awareness ? the knowledge one has about their current attentional state (e.g., Schooler, 2002) ? may be one strategy to improve task performance, and thus reduce falls, in seniors. Evidence from McVay and colleages (2009) suggests that performance declines when participants are unaware that they are currently mind-wandering (i.e., lack of ?meta-awareness?). Hence, training senior fallers to be more aware of their current attentional state may be an innovative way to improve performance and reduce falls among this at-risk population.  Our finding that the N200 was significantly modulated by attentional state, such that ERP amplitude was larger for mind-wandering blocks relative to on-task blocks was unexpected; previous ERP studies in mind-wandering have reported decreased amplitudes as a function of mind-wandering (Kam et al., 2011; Smallwood et al., 2008). Given the interpretation of the N200 as reflecting cognitive control (e.g., Folstein & Van Petten, 2008), our results support two main claims. First, although their thoughts have drifted off-task, older adults may still be prioritizing task performance on the SART. Indeed, older adults participating in such experiments tend to have higher motivation levels for completing the task well ? which may also contribute to overall lower mind-wandering rates in older adults compared to their younger counterparts. Second, increased N200 amplitudes may reflect a form of compensation for reduced cognitive resources associated with older age. Importantly, to understand the mechanisms behind why mean N200 amplitudes were larger in our sample of older adults during mind-wandering, future studies are required, including a comparison across the lifespan to examine whether this pattern of results may be age-dependent.   79 What might account for our failure to find a relationship between modulations in our ERP components during mind-wandering and falls? We speculate that while electrical activity may show normal patterns at the local level in fallers, abnormalities may be present in the connections between brain regions. The ?default mode network? (DMN) is a group of midline brain regions, including the posterior cingulate cortex, precuneous, and medial prefrontal cortex, and is active during the absence of an explicit external task (e.g., Raichle & Snyder, 2007). The connectivity ? or temporally synchronized activation ? within the DMN is greater during mind-wandering compared to being on-task (e.g., Mason et al., 2007; McKiernan et al., 2006). Recent work, however, has found that a disconnection in DMN activation is a hallmark characteristic of Alzheimer?s disease (e.g., Buckner, Andrews-Hanna, & Schacter, 2008; Greicius, Srivastava, Reiss, & Menon, 2004) ? which is associated with a two-fold increase in falls risk (Morris et al., 1987). In short, such disruptions within the DMN may provide further insight into the relationship between impaired cognition and falls.   A primary limitation of our study is that our classification of attentional states was based on subjective reports. As mentioned above, self-report is currently the ?standard? method for identifying when a person is engaging in task-unrelated thoughts. With increasing use of advanced technology in mind-wandering research, however, neural and physiological signatures may provide an objective measure of mind-wandering in future studies (e.g., Smallwood, in press). We highlight that previous work has found that mind-wandering frequency during laboratory-based tasks parallel mind-wandering rates in real-life within a given individual (McVay et al., 2009), providing ecological validation to our laboratory-based paradigm. Our work is also limited by the fact that we only included senior women in our study, due to the sample available to us. Further, we recognize that our sample is relatively small and homogeneous; thus, future work should focus on how the relationship between falls and mind-wandering may differ in a wider range of the population with larger sample sizes. Lastly, a caveat worth mentioning is that the benefits of attending  80 to the task at hand may be context-dependent, based on whether the focus is on internal versus external stimuli. In particular, over-focusing on postural control in older adults may be detrimental for sensory-motor performance.  To conclude, our study reveals an important link between falls and the frequency of engaging in task-unrelated thoughts. To the extent that mind-wandering during performance of an external task can be considered ?dual-tasking?, our results concur with current findings that fallers exhibit impaired dual-task performance, likely due to reduced processing capacity (e.g., Nagamatsu, Voss, et al., 2011; Rapport et al., 1998; Springer et al., 2006). Such reductions in processing capacity may also contribute to an inability to effectively prioritize tasks (e.g., Yogev-Seligmann et al., 2008) ? which also may account for our observed relationship between falls and frequency of mind-wandering. Our findings point towards the conclusion that older adults may engage in compensatory strategies ? including reduced frequency of mind-wandering and increased cognitive control during mind-wandering periods ? to account for decreased cognitive resources; such an inability to effectively engage in such compensatory behavior may thus contribute to falls. This is the first study to examine the frequency of mind-wandering and related neurocognitive processing changes as risk factors for falls. Future work should be aimed at elucidating the mechanisms by which mind-wandering may directly or indirectly impact the occurrence of falls. Finally, future research on how mind-wandering may contribute to falls in the context of the real-world ? such as through altered gait patterns ? would provide further insight into the relationship between falls and mind-wandering.     81 CHAPTER 5: Functional neural correlates of reduced physiological falls risk Introduction Falls are a major health care problem for seniors and health care systems. They are the third leading cause of chronic disability worldwide United Nations (2002) and approximately 30% of community-dwellers over the age of 65 years experience one or more falls every year (e.g., Tinetti et al., 1988). Importantly, 5% of falls result in fracture, with one-third of those being hip fractures.  Key risk factors for falls include reduced physiological function, such as impaired balance, (e.g., Lord, Clark, & Webster, 1991; Lord, Ward, Williams, & Anstey, 1994) and cognitive impairment (e.g., Tinetti et al., 1988). Recent evidence suggests that even mild reductions in cognitive abilities among otherwise healthy community-dwelling older adults increase physiological falls risk (e.g., Anstey et al., 2006; Liu-Ambrose, Ahamed, et al., 2008; Liu-Ambrose, Ashe, et al., 2008; Lundin-Olsson et al., 1997). Specifically, evidence suggests that reduced executive functions -- the ability to concentrate, to attend selectively, and to plan and strategize -- are associated with increased falls risk among seniors without cognitive impairment and dementia (e.g., Anstey et al., 2006; Holtzer et al., 2007; Lord & Fitzpatrick, 2001; Lundin-Olsson et al., 1997; Persad et al., 1995).  Currently, the neural basis for the association between reduced executive functions and falls is unclear. Evidence from neuroimaging studies provides insight to possible underlying mechanisms. Specifically, cerebral white matter lesions (or leukoaraiosis) are associated with both reduced executive functions (e.g., Thal, Del Tredici, & Braak, 2004) and gait and balance abnormalities (e.g., Baloh, Ying, & Jacobson, 2003; Briley, Wasay, Sergent, & Thomas, 1997; Masdeu et al., 1989; Soumare et al., 2009). Cerebral white  82 matter lesions may interrupt frontal lobe circuits responsible for normal gait and balance or they may interfere with long loop reflexes mediated by deep white matter sensory and motor tracts (e.g., Masdeu et al., 1989). In addition, the periventricular and subcortical distribution of white matter lesions could interrupt the descending motor fibers arising from medial cortical areas, which are important for lower extremity motor control (e.g., Baloh et al., 2003). However, while the results of these neuroimaging studies contribute to our appreciation of the importance of brain structure to physiological falls risk, they do not provide specific guidance for refining or developing falls prevention strategies because white matter lesions are not currently modifiable once they present. Studies have also demonstrated the contribution of brain volume to physiological falls risk. Specifically, reduced grey matter volume within sensorimotor and frontal parietal regions of the brain is associated with both reduced gait speed and impaired balance (e.g., Rosano et al., 2008; Rosano, Aizenstein, Studenski, & Newman, 2007).     Of particular relevance to falls prevention, targeted exercise training is beneficial for both brain volume as assessed by MRI and brain function as assessed by fMRI (e.g., Colcombe et al., 2006). What has not been well examined to date is the contribution of brain function to physiological falls risk. Using fMRI we previously demonstrated that reduced activity in the posterior lobe of the right cerebellum during an executive-challenging cognitive task may be an underlying neural mechanism for increased falls risk (Liu-Ambrose, Nagamatsu, et al., 2008).   To our knowledge, it is currently unknown whether the function of brain regions responsible for executive functions are independently associated with reduced physiological falls risk after accounting for relevant factors such as baseline age, baseline physiological falls risk, and baseline brain volume. Yet, such knowledge would facilitate the development and refinement of targeted interventions to reduce physiological falls risk in older adults. Thus, we used fMRI to examine the functional neural correlates of  83 executive functioning that are independently associated with reduced physiological falls risk among community-dwelling senior women.  Methods Participants The sample for this analysis consisted of a subset of 155 women who consented and completed a 12-month randomized controlled trial of exercise (NCT00426881) that primarily aimed to examine the effect of once-weekly or twice-weekly resistance training compared with a twice-weekly balance and tone exercise intervention on cognitive performance of executive functions. The design and the primary results of the study have been reported elsewhere (Liu-Ambrose et al., 2010).   We recruited and randomized 155 senior women who: 1) were aged 65-75 years; 2) were living independently in their own home; 3) obtained a score > 24 on the MMSE (Folstein et al., 1975); and 4) had a visual acuity of at least 20/40, with or without corrective lenses. We excluded those who: 1) had a diagnosed neurodegenerative disease (e.g., AD) and/or stroke; 2) were taking psychotropic drugs; 3) did not speak and understand English; 4) had moderate to significant impairment with ADLs as determined by interview; 5) were taking cholinesterase inhibitors within the last 12 months; 6) were taking anti-depressants within the last six months; or 7) were on oestrogen replacement therapy within the last 12 months.   Ethical approval was obtained from the Vancouver Coastal Health Research Institute and the University of British Columbia?s Clinical Research Ethics Board. All participants provided written informed consent.    84 Randomization The randomization sequence was generated by www.randomization.com and was concealed until interventions were assigned. This sequence was held independently and remotely by the Research Coordinator. Participants were enrolled and randomised by the Research Coordinator to one of three groups: once-weekly resistance training (1x RT), twice-weekly resistance training (2x RT), or twice-weekly balance and tone (BAT).    Exercise intervention Resistance Training All classes were 60 minutes in duration. The protocol for this program was progressive and high-intensity in nature. Both a Keiser? Pressurized Air system and free weights were used to provide the training stimulus. Other key strength exercises included mini-squats, mini-lunges, and lunge walks.   Balance and Tone This program consisted of stretching exercises, range of motion exercises, kegals, balance exercises, and relaxation techniques. This group served to control for confounding variables such as physical training received by traveling to the training centres, social interaction, and lifestyle changes secondary to study participation.  Descriptive variables Global cognition was assessed using the MMSE (Folstein et al., 1975). We used the 15-item GDS to screen for depression. FCI was calculated to estimate the degree of comorbidity associated with physical functioning (Charlson, Pompei, Ales, & MacKenzie, 1987). This scale?s score is the total number of comorbidities.  85 Dependent variable: physiological falls risk Physiological falls risk was assessed using the short form of the PPA (Prince of Wales Medical Research Institute, AUS) to assess physiological falls risk. The PPA measures five domains of physiological functioning ? dominant hand reaction time, postural sway, contrast sensitivity, proprioception, and dominant quadriceps strength ? and computes a global falls risk score that has 75% accuracy for predicting falls. Global PPA scores < 0 indicate low falls risk, 0 to 1 indicate mild falls risk, 1 to 2 indicate moderate falls risk, and scores > 2 indicate high falls risk. We calculated change in physiological falls risk as the difference score between the baseline global PPA score and the trial completion PPA score; higher PPA change scores indicate greater reductions in physiological falls risk.  Independent variables of interest Brain Structure: Anatomical MRI Baseline brain volume was measured via high-resolution, T1-weighted structural MRI images obtained using a Philips Achieva 3T scanner (TR = 8 ms, TE = 3.7 ms, bandwidth = 2.26 kHz, voxel size = 1 x 1 x 1 mm). Brain tissue volume, normalized for subject head size, was estimated with SIENAX (Smith et al., 2002), part of FSL (FMRIB?s Software Library, Version 4.1.4) (Smith et al., 2004). SIENAX starts by extracting brain and skull images from the single whole-head T1 image (Smith, 2002). The brain image was then affine-registered to Montreal Neurological Institute (MNI) 152 space (Jenkinson, 2003; Jenkinson & Smith, 2001). Next, tissue-type segmentation with partial volume estimation was carried out (Zhang, Brady, & Smith, 2001) in order to calculate baseline total volume of brain tissue, total white matter volume, and total grey matter volume.     86 Brain Function: Functional MRI Transverse echo-planar imaging (EPI) images in-plane with the AC-PC line were acquired using a gradient-echo pulse sequence and sequential slice acquisition (TR = 2000 ms, TE = 30 ms, flip angle = 90?, 36 contiguous slices at 3 mm skip 1 mm, in-plane resolution of 128 x 128 pixels reconstructed in a FOV of 240 mm). Each functional run began with four TR?s during which no data were acquired to allow for steady-state tissue magnetization. A total of 148 EPI volumes were collected in each functional run, and a total of six functional runs were collected for each participant.   During scanning, participants performed a modified Eriksen flanker task (e.g., Colcombe et al., 2004) -- a task that engages the executive cognitive processes of selective attention and conflict resolution (Figure 5.1). Participants viewed displays with an arrow at central fixation, flanked by a pair of arrows on either side. In half the trials, the flanking arrows pointed in the same direction as the central arrow cue (e.g., < < < < <; congruent condition), and in the other half, the flanking arrows pointed in the opposite direction (e.g., > > < > >; incongruent condition). There were four event types based on whether the central arrow was congruent versus incongruent with the distracter arrows and whether it pointed to the left or right. A central fixation cross was presented for 500 milliseconds at the beginning of each trial. The target stimulus (arrows) was then shown for 2000 milliseconds. An average of 13500 milliseconds of blank screen was presented between each trial, jittered between 11500 and 15000 milliseconds. Each participant underwent six successive five-minute blocks where they were presented with 17 trials that were first-order counterbalanced such that consistent and inconsistent trials followed each other equally. The participants' task on each trial was to signal the direction the central arrow points via a simple key press. Reaction time was recorded in milliseconds. At the end of the sessions, a high-resolution scan allowed each participant?s anatomical and functional images to be co-registered during data analysis.    87 Figure 5.1. The Flanker Task.  Participants were presented with a 13.5-sec fixation cross, which was followed by a 500 milliseconds pre-cue that informed participants that the critical stimulus will appear soon. Finally, an array of five arrows was on the screen. Participants responded to the orientation of the central arrow cue by pressing a button with their left hand if the central arrow cue pointed to the left and with their right hand if the central arrow cue pointed to the right. During one half of the trials, the flanking arrows faced in the same direction as the central arrow cue (i.e., congruent trials), and during the other half, they pointed in the opposite direction as the central arrow cue (i.e., incongruent trials). These stimuli remained on the screen for 2,000 milliseconds. Each participant underwent six successive five-minute blocks where they were presented with 17 trials that are first-order counterbalanced such that consistent and inconsistent trials followed each other equally (Colcombe et al., 2004). This paradigm is sensitive to age-related decrements in attention control (e.g., Kramer, Hahn, & Gopher, 1999).  Functional MRI data were processed and analyzed using SPM2 (http://www.fil.ion.ucl.ac.uk/spm). For each participant the EPI images were corrected for motion using the INRIalign toolbox for SPM2 (http://www-sop.inria.fr/epidaure/software/INRIAlign/). The resulting images were spatially-normalized into MNI stereotaxic coordinates using the EPI template provided with SPM2 (Friston, Ashburner, Frith, Poline, & Heather, 1995), and spatially smoothed using an isotropic 8 mm Gaussian kernel. For each participant, the smoothed, normalized EPI data were analyzed via multiple regression using a fixed-effects general linear model (Friston, Frith, Turner, & Frackowiak, 1995). In particular, the event-related responses to the onsets of the objects was examined, with each participant's model including four event-related regressors: 1. One for each combination of target type (i.e., left or right); and 2. Distracter condition (i.e., congruent or  88 incongruent). Regressors were based on the canonical event-related hemodynamic response function, temporal derivatives of the event-related responses were included as additional regressors, and low-frequency scanner and/or physiological noise was modeled via linear, quadratic, and cubic regressors of non-interest. Group-level analyses were then based on a random-effects model using one-sample t-tests, with a threshold of p < 0.05, corrected, and a minimum extent threshold of 10 contiguous voxels. Mean beta values reported for clusters identified in the group-level data were extracted from the SPM2 data files using custom scripts implemented in MATLAB (The MATHWORKS Inc., Natick, MA). The group-level cluster means were calculated by first determining each participant?s mean beta across all voxels in the given cluster. All reported voxel coordinates were converted to Talairaich coordinates (Talairach & Tournoux, 1988)  using a modified version of the mni2tal MATLAB script (www.harvard.edu/~slotnick/scripts.htm). The mean beta values were then imported to SPSS.    Statistical analyses Descriptive data are reported for variables of interest. Data were analyzed using SPSS Windows Version 18.0 (SPSS Inc., Chicago, IL). The associations between the variables were determined using the Pearson product moment coefficient of correlation.   A multiple linear regression model was constructed to determine the independent association of the neural correlates of executive functioning, as assessed by fMRI, with change in physiological falls risk over the 12-month intervention study, as assessed by PPA. Baseline age, experimental group, and baseline physiological falls risk were statistically controlled by entering these three variables into the regression model first. These independent variables were determined from the results of the Pearson product moment coefficient of correlation analyses (i.e., baseline PPA score) and based on biological relevance, such as experimental group and age.   89  Baseline total brain volume, total white matter volume, and total grey matter volume were then entered into regression model and only those that significantly improved the model were included (i.e., stepwise). Finally, regions of the brain (i.e., clusters) showing increases in the hemodynamic response on incongruent relative to congruent trials of the flanker task were then entered into the model using a stepwise approach. Alpha was set at p < 0.05.  Results Participants and variables of interest Of the 155 participants who consented and were randomized at baseline, 135 completed the 12-month trial. Seventy-three of the 135 participants consented and completed baseline MRI and fMRI scanning.   Table 5.1 reports the baseline descriptive statistics for this cohort. The mean baseline PPA score was 0.10, indicating mild falls risk. At the end of the 12-month trial, the 73 women demonstrated a mean change of 0.10 in the PPA score.  A paired t-test indicated that this was not a statistically significant change (p = 0.06).      90 Table 5.1. Descriptive statistics for variables of interest (N=73).  1BAT = Balance and Tone; 1x RT = once-weekly resistance training; 2x RT = twice-weekly resistance training; yr = year; kg = kilogram; MMSE = Mini-Mental State Examination; sec = seconds. 2Count = number of ?yes? cases within each group. % = percent of ?yes? within each group.  Variable1  BAT  (n=22)  1x RT (n=28) 2x RT (n=23) Total  (N=73)  Mean (SD) Mean (SD) Mean (SD) Mean (SD) Age (yr) 69.6 (3.1) 69.5 (2.7) 69.1 (3.1) 69.4 (2.9) Height (cm) 161.5 (6.2) 162.0 (7.5) 162.4 (6.9) 161.9 (6.9) Weight (kg) 67.1 (10.9) 67.9 (13.6) 68.6 (13.0) 67.9 (12.5) Education2     Less than Grade 9 0.0 (0.0) 0.0 (0.0) 0.0 (0.0) 0.0 (0.0)   Grade 9 to 12 without Certificate or Diploma 2.0 (9.1) 2.0 (7.1) 0.0 (0.0) 4.0 (5.5) High School Certificate or Diploma 5.0 (22.7) 3.0 (10.7) 5.0 (21.7) 13.0 (17.8)   Trades or Professional Certificate or Diploma 3.0 (13.6) 6.0 (21.4) 2.0 (8.7) 11.0 (15.1) University Certificate or Diploma 4.0 (18.2) 5.0 (17.9) 4.0 (17.4) 13.0 (17.8) University Degree 8.0 (36.4) 12.0 (42.9) 12.0 (52.2) 32.0 (43.8) MMSE Score (max. 30 pts) 28.8 (1.3) 28.6 (1.3) 28.8 (1.0) 28.7 (1.2) Falls in the Last 12 Months (yes/no) 8 (36.4) 7 (25.0) 9 (39.1) 24 (32.9) GDS (/15 pts) 0.7 (2.2) 0.1 (0.8) 0.6 (1.6) 0.5 (1.6) FCI (/18 pts) 2.2 (1.3) 1.9 (1.7) 1.7 (1.5) 1.9 (1.5) Baseline PPA Score 0.10 (0.91) 0.06 (0.89) 0.16 (1.11) 0.10 (0.96) Total Brain Volume (mm3) 1404767.07 (61101.38)  1392824.85 (74770.29) 1425571.35 (53607.47) 1406741.26 (65216.88) White Matter Volume (mm3) 673259.09 (37763.87) 668611.61 (33667.89) 680775.90 (30457.92) 673844.81 (33920.23) Gray Matter Volume (mm3) 731508.20 (30004.57) 731957.93 (35834.91) 746535.05 (35339.77) 736415.19 (34256.96) Change in Physiological Falls Risk 0.25 (0.97) 0.04 (0.88) 0.34 (0.82) 0.10 (0.96)  91 Behavioural performance on the flanker task was calculated as percent increase in reaction time to incongruent stimuli, over and above the average reaction time to congruent stimuli [[(incongruent reaction time ? congruent reaction time) / congruent reaction time] x 100] (e.g., Colcombe et al., 2004). The percent increase measure is derived to reflect interference unbiased by differences in base reaction time. Only correct responses were included in the analysis. Mean interference score for BAT, RT1, and RT2 were 16.59 (SD = 13.07), 19.92 (SD = 2.52), and 27.98 (SD = 13.77), respectively.  Consistent with previous studies using the flanker task, regions showing increases in the hemodynamic response on incongruent relative to congruent trials included bilateral inferior and middle frontal gyri, frontal orbital cortex, anterior cingulate cortex (ACC), bilateral precuneus, and the right cerebellum; 14 clusters were identified (Table 5.2). 92 Table 5.2. Voxel Cluster Statistics from fMRI.  Hemisphere Structure BA1 K2 t3 MNI TAL      X Y Z X Y Z Right Lateral occipital cortex 19 4247 9.51 28 -78 42 28 -74 42 Right Frontal orbital cortex 47 597 8.18 36 24 -4 36 23 -5 Right Posterior cerebellum  1131 7.95 8 -80 -34 8 -79 -25 Left Lateral occipital cortex 37 626 7.67 -48 -70 -12 -48 -68 -7 Right Paracingulate gyrus 32 1634 7.49 8 20 44 8 21 39 Left Lateral occipital cortex 7 2915 7.43 -22 -72 32 -22 -68 33 Left Middle frontal gyrus 6 1620 7.20 -26 0 50 -26 2 46 Right Middle frontal gyrus 6 631 7.03 26 2 48 26 4 44 Right Inferior frontal gyrus 9 699 6.97 54 14 28 53 15 25 Left Frontal orbital cortex 47 198 6.51 -32 24 -6 -32 23 -6 Left Frontal orbital cortex 47 122 6.19 -46 20 -10 -46 19 -9 Right Supramarginal gyrus 40 82 6.04 28 -68 -28 28 -67 -20 Right Posterior cerebellum  25 5.72 14 -76 -50 14 -76 -38 Right Anterior cerebellum  21 5.69 16 -38 -34 16 -38 -27 Reported coordinates and t values are for the cluster maxima. 1BA = Brodmann?s area.  2 K = # of voxels in the cluster.  3 All p values < 0.05.   Data are group-averaged across all 83 participants and shown on a rendered brain provided with SPM2.  Data were thresholded at p < 0.05 (corrected) and a minimum cluster size of 10 contiguous voxels. The left  93 OFC-In and right PCG-ACC both contributed significantly to our model predicting change in physiological falls risk (Figure 5.2).   Figure 5.2. Brain regions demonstrating an increased hemodynamic response on incongruent relative to congruent trials.      94 Correlation coefficients Table 5.3 reports the bivariate correlation coefficients of those variables included in the final multiple linear regression model. Baseline physiological falls risk was positively and significantly associated with change in physiological falls risk (p < 0.001). Baseline total brain volume, total white matter volume, and activation (i.e., hemodynamic response) in the left frontal orbital cortex extending towards the insula (OFC-In) were negatively and significantly associated with change in physiological falls risk (p < 0.05). In our bivariate analysis, age, experimental group, and activation in the right paracingulate gyrus extending towards the anterior cingulate cortex (PCG-ACC) were not associated with change in physiological falls risk (p > 0.26).        95 Table 5.3. Multiple linear regression model summary for improved physiological falls risk.   ? PPA Score (Baseline Score ? Trial Completion Score)  Independent Variable r R2 R2 Change Unstandardized B (Standard Error) Standardized ? p - value Model 1 0.565 0.319 0.319    Group 0.040   0.015 (0.112) 0.013 0.896 Age  -0.078   -0.064 (0.031) -0.211 0.043 Baseline PPA Score 0.526**   0.529 (0.094) 0.575 <0.001 Model 2 0.619 0.383 0.064       Group 0.040   0.040 (0.107) 0.035 0.713 Age  -0.078   -0.068 (0.030) -0.224 0.026 Baseline PPA Score 0.526**   0.521 (0.090) 0.566 <0.001 White Matter Volume -0.263*   -6.670E-6 (0.000) -0.255 0.010 Model 3  0.698 0.487 0.104       Group 0.040   0.034 (0.099) 0.030 0.733 Age  -0.078   -0.088 (0.028) -0.287 0.003 Baseline PPA Score 0.526**   0.504 (0.083) 0.548 <0.001 White Matter Volume -0.263*   -8.800E-6 (0.000) -0.337 <0.001 Cluster 31 -0.258   -0.654 (0.177) -0.339 0.014 Model 4 0.729 0.531 0.044       Group 0.040   0.023 (0.095) 0.021 0.809 Age  -0.078   -0.087 (0.027) -0.286 0.002 Baseline PPA Score 0.526**   0.474 (0.081) 0.515 <0.001 White Matter Volume -0.263*   -1.000E-5 (0.000) -0.383 <0.001 Cluster 31 -0.258*   -1.159 (0.266) -0.601 <0.001 Cluster 72 -0.055   0.637 (0.271) 0.329 0.016 * p ? 0.05 **p ? 0.001 1Cluster 3 is the region of left frontal orbital cortex extending towards the insula. 2Cluster 7 is the region of right paracingulate gyrus extending towards the anterior cingulate cortex.   Linear regression model Baseline age, experimental group, and baseline physiological falls risk, accounted for 31.9% of the variance in change in physiological falls risk (Table 5.3). Adding baseline total white matter volume resulted  96 in an R-square change of 6.4% and significantly improved the regression model (F Change = 7.1, p = 0.01). Adding activation in the left OFC-In to the model resulted in an R-square change of 10.4% and significantly improved the model (F Change = 13.6, p < 0.001). Finally, the inclusion of activation in the right PCG-ACC resulted in significant R-square change of 4.4% (F Change = 6.6, p = 0.02). The total variance accounted by the final model was 53.1% (Table 5.3). Based on the standardized betas, the left OFC-In was most associated with reduced physiological falls risk.  Discussion Recent evidence strongly suggests that changes in brain structure with age contribute to problems with mobility (e.g., Rosano et al., 2008; Rosano, Aizenstein, et al., 2007; Rosano, Brach, Longstreth Jr, & Newman, 2006; Rosano, Brach, Studenski, Longstreth, & Newman, 2007; Rosano et al., 2005). However, less is known about the role of brain function (e.g., Liu-Ambrose, Nagamatsu, et al., 2008). To our knowledge, our study is the first to demonstrate the independent contribution of brain function to reduced physiological falls risk among community-dwelling seniors. Specifically, after accounting for baseline age, experimental group, baseline physiological falls risk, and baseline total white matter volume, activation in the left OFC-In was negatively and independently associated with reduced physiological falls risk in community-dwelling senior women over a 12-month period. In contrast, activation in the PCG-ACC was positively and independently associated with reduced physiological falls risk.  The two regions included in our multiple linear regression model -- the left OFC-In and the right PCG-ACC -- are both part of the neural network associated with response inhibition and selective attention (e.g., Boehler, Appelbaum, Krebs, Hopf, & Woldorff, 2010; Carter et al., 1998; Casey et al., 1997; Haupt, Axmacher, Cohen, Elger, & Fell, 2009). Response inhibition ? the ability to avoid unwanted, inappropriate responses ? is associated with falls in seniors. For example, Anstey and colleagues (2009) reported that  97 senior fallers (both single and recurrent) performed significantly worse on a measure of response inhibition compared to non-fallers. The authors suggested that reduced inhibition results from age-related declines in functioning of the prefrontal cortex, which contributes to falls. Given that movement through the environment requires attending to relevant stimuli and inhibiting prepotent, yet potentially unsafe, responses, it is not surprising that brain regions associated with response inhibition and selective attention are related to falls risk.  Importantly, we found that activation in the left OFC-In was negatively associated with reduced physiological falls risk, whereas activation in the PCG-ACC was positively associated with reduced physiological falls risk. Increased activation in the frontal cortex during an executive task, such as the flanker, is associated with better task performance (e.g., Colcombe et al., 2004). In contrast, increased activation of the anterior cingulate cortex in older adults is associated with reduced task performance (e.g., Colcombe et al., 2004). In particular, increased ACC activation is hypothesized to be an indicator of greater cognitive effort such that the ACC is less efficient at triggering the prefrontal system to engage cognitive control (e.g., Botvinick, 2007).   Our volumetric brain results also suggest that total white matter volume, rather than total grey matter volume, is associated with change in physiological falls risk. Previous research suggests that white matter declines at a faster rate than grey matter in otherwise healthy older adults (e.g., Resnick, Pham, Kraut, Zonderman, & Davatzikos, 2003). Our results extends this finding by suggesting the loss of total white matter volume may be an early indicator of increased falls risk among community-dwelling older adults.  Of particular clinical relevance, the results of our study suggest that individuals at higher risk for future falls have greater potential for risk reduction than those at lower risk for falls. Specifically, our multiple  98 regression model showed that baseline physiological falls risk was positively associated with change in physiological falls risk. Hence, our current study results concurs and extends that of a previous meta-analysis that concluded exercise-based falls prevention strategies are most effective among those at the greatest risk (e.g., Robertson, Campbell, Gardner, & Devlin, 2002). This suggests that one intervention strategy for falls prevention may be to target those who are at greatest risk for falls.  We note that of the independent variables included in our regression model, baseline activation of the left OFC-In was most associated with reduced physiological falls risk. Hence, while many falls interventions focus on balance training, our study suggests that future falls prevention strategies should potentially incorporate intervention components that induce neurocognitive plasticity (i.e., changes in brain function). Future work is needed to establish whether such interventions would be effective. Current evidence suggests that targeted aerobic exercise training has specific benefits on neurocognitive plasticity in brain regions that are responsible for selective attention and response inhibition (e.g., Colcombe et al., 2004). Therefore, promoting plasticity in brain regions associated these key executive functions, may have a positive impact on falls prevention.  We acknowledge that our finding of a negative association between baseline total white matter volume and change in physiological falls risk is significantly associated with reduced falls risk contrasts previous cross-sectional studies on gray matter volume, balance, and mobility. Specifically, Rosano and colleagues (2008; 2007) found that reduced gait speed and impaired balance ? key risk factors for falls -- were significantly correlated with reduced grey matter volume within sensorimotor and frontal parietal regions in the brain. However, we highlight that our study examined the independent contribution of baseline volumetric brain measures to change in falls risk (i.e., longitudinal study design versus cross-sectional design) and hence  99 our conclusion that those at the greater risk for future falls (i.e., smaller baseline total white volume) have greater potential for falls risk reduction (i.e., greater change in PPA scores).  We recognize the limitations of our study. A key limitation is that we did not quantify white matter lesions within total white matter volume. We note that our study sample consisted exclusively of independent community-dwelling senior women who were without significant physical and cognitive impairments and without a significant history of falls. Thus, the results of our study may not generalize beyond this population of senior women and we may have underestimated the contribution of brain function to change in physiological falls risk. Future prospective studies are needed to test whether the present findings also apply to larger, more heterogeneous populations.  To conclude, the function of brain regions underlying response inhibition and selective attention was independently associated with reduced physiological falls risk. Hence, future falls prevention strategies should potentially incorporate intervention components, such as aerobic exercise training, that induce neurocognitive plasticity in the neural network that supports response inhibition and selective attention.    100 CHAPTER 6: General discussion In this dissertation, I have provided converging evidence from four separate studies leading to the overarching conclusion that impaired attentional processing is associated with falls/falls risk in older adults. This chapter is aimed at providing a general discussion to summarize and integrate the information presented thus far. I will begin with a synopsis of the studies included in my dissertation and revisit the research questions set forth at the beginning of this dissertation. A discussion of the limitations of my research and research in the field of falls and cognition, broadly speaking, will follow. I will conclude with potential future directions for this increasingly important area of research.  Summary of Studies  In Chapter 2, I examined how reduced sensory processing of peripheral probes, as measured via the N1 ERP component, was associated with falls risk. I found that reduced sensory processing in the left visual field, measured at ipsilateral electrode sites, specifically, was associated with falls risk. This study points towards right hemispheric and/or inter-hemispheric transfer problems, and is paramount for elucidating underlying structural and/or functional issues in the brain that may contribute to falls risk. In Chapter 3, I found that increased cognitive load impaired decision making/judgments in a virtual reality task in older adults at-risk for falls. Importantly, this study shows that falls risk is associated with reduced dual-task performance in a real-life situation; furthermore, these results demonstrate that dual-task performance does not just negatively impact physical abilities, but extends to higher-level cognitive processes as well. The study in Chapter 4 was designed to examine whether falls might be associated with an increased tendency to mind-wander. I found that the number of falls recently experienced was positively correlated with mind-wandering frequency. My findings from this study suggest that falls may be associated with an impaired ability to appropriately allocate attentional resources to the task at hand. Finally, Chapter 5 provides novel  101 evidence that two regions in cortex are functionally associated with falls risk ? the left OFC-I and the right PCG-ACC. This result provides an essential basis for understanding underlying neural mechanisms for falls and guides future research on understanding connections between the brain, cognition, and falls.  Main Research Questions Revisited 1. What specific types of attention are associated with falls/falls risk? The research that comprises my dissertation provides evidence that fallers do indeed have impaired attentional processing. To briefly recapitulate, my research suggests that falls and falls risk are associated with reduced: 1. Attentional orienting to task-irrelevant information; 2. Attention to a primary task under dual-task conditions ? including mind-wandering; and 3. Functional activation in neural networks that subserve selective attention and response inhibition. While these distinct domains of attention that have now been linked to falls may seem divergent and independent from each other, my proposed framework below will provide a unified way to think about the specific deficits in attention that are experienced as a function of falls/falls risk.  Connecting my obtained results with current theories on attention, executive cognitive functions, and falls, I propose that the relationship between attention and falls risk can be modeled by an attentional resource framework. In particular, two dissociable ? but often co-occurring ? factors appear to play an integral role in falls risk: 1. Limited attentional resources to dedicate to ongoing task(s); and 2. Difficulty with appropriately allocating resources to the most pertinent task.  Notably, ?resources? represent a broad and multi-domain concept; accordingly, they can be defined in multiple ways. In the context of my data, resources can be construed in at least two independent ways. First, we can think of perceptual resources, where spatial attention can be considered a deployable  102 resource. This conceptualization of resources was presented in Chapter 2, where participants engaged in a primary task at fixation while peripheral task-irrelevant probes were presented on the left and right side of visual space. Falls risk was associated with reduced sensory processing of left probes, suggesting that fewer perceptual resources were being directed to the left periphery. This corroborates previous reports that fallers have a reduced ability to attend to the left side of space during a spatial orienting task (Nagamatsu et al., 2009).  Second, attentional resources can be considered at the ?task? level, where attention can be allocated to one task over another under dual-task situations. In this regard, we know that if cognitive demand outweighs available resources, task performance will deteriorate (e.g., Kahneman, 1973). As discussed extensively throughout this dissertation, it has been established that fallers exhibit higher DTC during the performance of two tasks simultaneously, providing support for the notion that fallers have reduced attentional resources. Specifically, in Chapter 3, older adults at-risk for falls had impaired performance on the simulated street-crossing task while they were concurrently engaging in a conversation on a phone. Notably, such decrements to performance were not observed in the control condition, or the cognitively easier task requiring them to listen to music on an iPod. Thus, it appears that only under high cognitive load, those at-risk for falls have fewer resources to dedicate to a secondary task.  Interestingly, Chapter 4 on mind-wandering and falls history examines both of the above types of resources. In that study specifically, I examined how periods of mind-wandering can alter the perceptual and cognitive processing of visual information, in addition to whether attention was focused on the task at-hand versus task-unrelated thoughts at any given moment in time. In this study, I did not find any relationship between falls and resources at the perceptual level, which may be attributed to the perceptual easiness of the task. In particular, because there was no manipulation of perceptual load, differences  103 between fallers and non-fallers were not observed. In contrast, fallers did show an increased frequency of mind-wandering, demonstrating an inability to effectively allocate their task-related resources while inhibiting attention towards task-irrelevant information (e.g., inner thoughts).  One additional point worthy of discussion regarding attentional resources as applied to the data I have presented in this dissertation, is whether my findings provide incongruent results. Specifically, while the results of Chapter 2 suggest that falls risk is associated with increased focus on the central task, while ignoring task-irrelevant peripheral stimuli, Chapter 4 provides evidence that fallers have reduced attention to the primary task, while engaging in task-unrelated thoughts. Within the context of the framework presented above ? that ?resources? can be represented in independent ways ? I argue that my results converge on the notion that falls risk may be associated with impaired resource quantity and allocation at multiple levels. That is, these types of attentional resources are not mutually exclusive, such that one may exhibit an impaired ability to direct resources at a perceptual level, in addition to problems with effectively allocating task-level attention. Importantly, regardless of whether the deficit manifests as increased or decreased attention to the primary task, such problems with the control of attentional resources may have negative behavioural consequences. The specific alteration in engagement to a primary task (i.e., whether it?s increased or decreased) may be situation-dependent; further, another possibility is that fallers may not only experience impairments in where attention is directed (task-related vs. unrelated information), but also in the actual switching between tasks (discussed in more detail below under future directions).  2. How might impaired attentional processing contribute to increased falls risk? One proposed way in which attentional processing and falls risk are linked is via postural control. We know that executive functioning is a critical component of balance and gait, and that postural stability is even more dependent on cognitive processes with age (e.g., Lindenberger et al., 2000; Lovden et al., 2008;  104 Wollacott & Shumway-Cook, 2002; Yogev-Seligmann et al., 2008). Specifically, postural stability ? the ability to maintain the position of the body within specific boundaries of space (Lord et al., 2007) ? requires an interaction between musculo-skeletal and sensory systems, in order to integrate information about body position in space and generate forces to control body movement (Lord, et al., 2007). While these two systems work in consort, they are integrated by higher-level cognitive processes in order to seamlessly plan movements and successfully adapt to changing environmental demands (Lord, et al., 2007). An inability to cope with such demands may result in instability, poor motor control, and consequently, falls.   Importantly, postural control requires cognitive resources, and an allocation of cognitive resources to focus on physiological integrity ? both of which were discussed above as being impaired as a function of falls risk. Impaired executive functioning may lead to a reduced reserve of cognitive resources available to perform multiple tasks simultaneously. In addition, an inability to appropriately allocate resources to the postural task may also place individuals at increased physiological falls risk. Indeed, seniors, and especially those with neurodegenerative disease, may not employ the typical ?posture-first? strategy necessary to prioritize gait over secondary task performance (e.g., Bloem, Grimbergen, van Dijk, & Munneke, 2006); Holmes, et al., 2010). Thus, an inability to allocate adequate resources to the postural task, in favour of carrying on a conversation, for example, may contribute to falls. Thus, it appears that impaired attentional processing may lead to falls indirectly, with balance and mobility providing a critical link between the two.   A second proposed way in which impaired attentional processing may lead to falls is via visuo-motor abilities, which are critical for our ability to effectively plan our movements through the environment while avoiding relevant hazards. Converging evidence has suggested that falls risk may be linked to reduced sensory-perceptual processing to the left side of visual space (e.g., Chapter 2; Nagamatsu, Carolan, Liu-Ambrose, & Handy, 2009). Importantly, such problems with the allocation of perceptual resources may  105 have negative real-life consequences, such as a failure to notice steps and curbs during navigation. Furthermore, a lack of attention to the environment could also mean that objects important for falls-avoidance could be ignored, such as handrails. Worth mentioning, these two proposed models for how impaired attentional processing may lead to increased falls risk are not mutually exclusive; indeed, a combination of factors may prove to be the most potent predictor of falls. Future research to examine the potential causal contribution of impaired attention to falls risk is warranted to examine the independent and interacting contribution of both postural control and visuo-motor abilities.  3. What underlying neural structures are implicated in the relationship between attention and falls risk? Chapter 5 in my dissertation provides evidence that two areas key for selective attention and response inhibition are associated with falls risk in older adults: the left OFC-I and the right PCG-ACC. It has been demonstrated that these regions are structurally and functionally connected (e.g., Menon & Uddin, 2010; Seeley et al., 2007). Menon and Uddin (2010) have proposed that the anterior insula and ACC are core components of the saliency network, which facilitates network switching between the central-executive and default-mode networks. Working in consort, the anterior insula first acts to detect salient sensory stimuli in the environment, then initiates control signals that engage higher-order cognitive processes while simultaneously disengaging the default-mode network (Menon & Uddin, 2010). The ACC is then critical for modulating motor output, given its direct connections to the spinal cord (Menon & Uddin, 2010).  The theory proposed by Menon and Uddin (2010) accompanied by the information provided above regarding attentional resources and falls, suggests that falls risk may be associated with problems selecting relevant information from the environment along with issues related to attentional control. First, in Chapter 2, I provided evidence that falls risk is associated with an impaired ability to attend to stimuli presented on  106 the left side of visual space. While it may be argued that sensory processing of task-irrelevant peripheral information is not required, in the context of the real world environment during navigation, being vigilant to any potential hazards ? regardless of the primary task ? is critical. Indeed, a narrowed spotlight of attention is associated with parietal damage (e.g., Townsend & Courchesne, 1994) and neuropsychological disorders such as autism (e.g., Travers, Klinger, & Klinger, 2011). Second, Chapters 3 and 4 suggest that falls/falls risk is associated with an impaired ability to direct adequate attention to a primary task to maintain behavioural performance. Interestingly, recent work has found that mind-wandering is linked with increased functional activation in both the central-executive and default-mode networks (e.g., Christoff et al., 2009). Therefore, falls risk may reflect an impaired ability to control attention towards a primary task while inhibiting default mode network activation. An important caveat to my research and resulting theories is that the work presented in my dissertation is correlational; hence, causal mechanisms between attention and falls risk remain speculative and future research is mandatory.  Limitations The limitations for each study included in my dissertation have been discussed within each relevant chapter. However, one over-arching limitation worth mentioning is that the studies in my dissertation include relatively small sample sizes and use homogenous samples (i.e., cognitively-healthy and community-dwelling women). Thus, the evidence that I have provided regarding the link between attentional processing and falls risk must be considered as preliminary; future research with larger sample sizes is required to further explore this relationship and to investigate whether my results extend to more diverse populations.  There are also several noteworthy limitations in research on falls and cognition in general, which I will outline here. First, there is currently no standardized metric used to classify falls status. For instance,  107 researchers can define falls status based on number of falls either prospectively or retrospectively. Alternatively, falls status can also be based on a measure of falls risk, such as physiological function. The result of using different classification schemes is that each study may use a different population, rendering it difficult to make cross-study comparisons. Furthermore, some classification methods may be more reliable than others. Self-reports of prior behaviour, for example, are subject to retrospective bias due to reduced memory capabilities in seniors. This problem is exacerbated in those with neurodegenerative disease (e.g., Bloem, Valkenburg, Slabbekoorn, & van Dijk, 2001). Thus, it would be useful to establish a standardized classification scheme that can be implemented universally.  Second, some limitations of research on cognition and falls have arisen because the majority of investigators conducting the research are not experts in cognitive psychology. Rather, they are medical professionals, physical therapists, and gerontologists. This results in a few key issues. To begin with, cognitive constructs, such as attention and executive cognitive functioning, are poorly defined, and are often based on lay understandings of the terms instead of being rooted in psychological theory. In addition, many of the studies use neuropsychological batteries to test for differences between fallers and non-fallers, hoping to see a significant difference on at least one measure, By giving participants a large number of tests at once, the chances of making a Type I error ? detecting a difference spuriously ? increases. In addition, most neuropsychological tests do not measure specific/individual constructs, making it difficult to determine the exact cognitive impairment that senior fallers may have. Fortunately, problems with using neuropsychological batteries are somewhat resolved by a large number of studies converging on the same story ? that senior fallers have problems with executive cognitive functioning. Furthermore, the dual-task paradigms that are commonly used tend to assess the physical consequences of DTC, rather than also measuring the corresponding impact on cognition. Overall, research on cognition and falls would benefit  108 from future collaborations across fields, where cognitive psychologists can work together with physical therapists and gerontologists.  A third limitation is the inability to directly examine the causal relationship between cognition and falls. Most studies to date have used prospective or cross-sectional designs to look at associations between falls and cognitive function. Because the occurrence of falls cannot be experimentally manipulated, intervention studies may be a practical resolution to this problem. For example, a cognitive training intervention which reduces falls may provide evidence that cognitive functioning does indeed impact the incidence of falls.  Last, an important consideration for future falls research is the issue of ecological validity. While studies have reported impairments on standard tests of executive cognitive functioning among seniors, these results do not always apply to real-world situations. Examining falls in the context of the real-world is especially critical, given that their causes are multi-factorial ? multiple risk factors are often involved, and the context of the situation plays an important role. Studying everyday behaviours, however, has inherent limitations as well. In particular, it does not allow distinct cognitive processes to be studied in isolation. One solution to the problem of internal versus external validity trade-off is to use a multi-method approach. Specifically, if both tests in the laboratory and in the field converge on the same result, then we can be more confident in our conclusions.  Future Directions 1. What accounts for the visual-field asymmetries previously found in fallers? Based on my previous research (Nagamatsu et al., 2009) and the work presented in Chapter 2, fallers appear to have specific deficits attending to stimuli in the left visual field. The brain processes information contralaterally, such that the right hemisphere is responsible for responding to information in the left visual  109 field (e.g., Mangun et al., 1994); this suggests that impairments observed in fallers may originate from deficits in the right hemisphere. I am currently investigating this possibility using an obstacle avoidance task that requires participants to reach towards a target in front of them while they ignore obstacles placed on the left and right side of the table. Using a motion tracking system, I am able to compare reach trajectories between fallers and non-fallers. Preliminary evidence shows that fallers show a rightward bias in their reach trajectories, suggestive of a mild form of visual neglect in those with a history of falls. Given that the right parietal lobe is implicated in unilateral visual neglect (e.g., Halligan et al., 2003), this study may provide further evidence for underlying mechanisms linking attention and falls.  2. What underlying neural circuits, structures, and functions contribute to falls risk? There are now multiple proposed functional regions and circuits that may contribute to falls risk. First, previous work conducted by myself and colleagues (Liu-Ambrose, Nagamatsu, et al., 2008) suggests that the cerebellum ? which is involved in motor control and connects with prefrontal cortex (e.g., Krienen & Buckner, 2009) ? is involved in falls. Second, my research in Chapter 5 provides evidence that two neural regions critical for attention are involved in falls risk. What is now required is a comprehensive understanding of how these regions may integrate to contribute to falls. Current research is being conducted in my laboratory to examine differences in functional connectivity ? both between and within large-scale brain networks ? between fallers and non-fallers (Hsu et al., in prep.).  3. Might impaired task-switching among fallers account for their observed attentional processing deficits? In addition to the resource framework proposed above to account for our findings regarding the relationship between attention and falls risk, my data can also be characterized by the idea of fallers exhibiting a ?sticky switch? when they are required to shift between tasks. Importantly, effective task-switching requires both  110 the ability to disengage from the first task as well as to redirect attention to the subsequent task. In Chapter 2 of my dissertation, falls risk was associated with reduced sensory processing of external stimuli, providing evidence of problems disengaging from the central stimuli. Furthermore, Chapter 4 demonstrated that older adults with a history of falls tend to mind-wander more frequently, contributing to the notion that fallers may have difficulty switching between attentional states. While fallers are known to demonstrate impaired performance on the Trail Making Test (e.g., Lord & Fitzpatrick, 2001), a test of set-shifting within a given task, the question of whether fallers might also be impaired on their ability to switch between tasks remains open ? and is an area of research worth future investigation.  4. Can we develop a simple and cost-effective test to evaluate executive cognitive function as a falls risk-factor? There is currently no widely-accepted, easy to administer, and reliable tool for assessing executive functioning among senior fallers. The MMSE is often used as a general screening tool for cognitive impairment. However, we now know that more specific domains of cognition ? and attentional processing ? may be critical for falls. A second widely-used tool is the Walking While Talking test. One limitation of the WWT, however, is that it requires space to administer, which may not always be available in crowded hospitals and busy clinics. The WWT is also subject to tester variability, such that reaction times may not be accurately measured or instructions may not be uniform.   An alternative tool may be a computerized version of a dual-task test. In Chapter 3, I provide evidence that performance on a simple dual-task test on a computer was highly correlated with walking and talking dual-task behaviour. Indeed, both results and accuracy on the tasks were matched. Furthermore, significant differences on dual-task performance were observed between older adults at-risk and those not-at-risk for falls. Thus, the computerized version may be a more objective and simple way to assess dual-task  111 performance. Future studies should investigate whether such a test might validly and reliably predict falls in a larger sample.  5. How can we use the information we?ve gained to reduce falls risk? The next step in research on attention and falls risk is to use the information we have gained to design and implement intervention strategies that can be applied in patients to reduce falls. Given what we now know, what would be the best type of intervention? Specifically, what type of intervention strategy would result in the lowest falls rate, while balancing cost and time effectiveness?  Cognitive training: Research on cognitive training in the past has been somewhat unsuccessful. In general, benefits of cognitive training do not transfer to other tasks (e.g., Ball et al., 2002). That is, training on one particular cognitive task improves performance on that task only, without extending to other tasks within the same or different cognitive domain. Therefore, before an effective cognitive training intervention can be developed, we need to identify a task that transfers to other domains. Research at the University of Illinois has focused on developing a video game training protocol to improve cognitive functioning (e.g., Ball et al., 2002). This is one line of research that I plan to pursue during my postdoctoral training.  Dual-task training: Interventions intended to improve dual-task performance have provided encouraging results. For example, in a study of young adults, the author found that training participants to dual-task was more effective in reducing DTC than training each task separately (e.g., balance and cognition) (e.g., Pellecchia, 2005). Similar findings have been reported in a sample of balance-impaired seniors (e.g., Silsupadol et al., 2009). In addition, there is evidence that cognitive dual-task training can improve balance and motor dual-task performance (e.g., Li et al., 2010) in seniors. An emerging line of research that has been made possible due to technological advances is VRE training. Mirelman and colleagues (2011) had  112 PD patients complete six weeks of VRE training by walking on a treadmill while simultaneously completing cognitive tasks. After the intervention, participants had improved gait and cognitive function, in addition to reduced DTC. Moreover, the benefits of the intervention persisted after a 4-week follow- up period. Importantly, improvements were observed beyond the tasks that were specifically trained ? i.e., transfer effects were obtained.  Exercise training: Randomized controlled trials (RCTs) of exercise have previously been found to improve executive cognitive function (e.g., Colcombe et al., 2004; Liu-Ambrose et al., 2010; Nagamatsu, Handy, Hsu, Voss, & Liu-Ambrose, 2012) and reduce falls (e.g., Liu-Ambrose, Donaldson, et al., 2008). Furthermore, exercise may decrease falls risk by improving cognitive ? rather than physiological ? functioning. In particular, an RCT investigated the effectiveness of the Otago Exercise Program, a resistance training and balance program designed to reduce falls, on improving both physiological and cognitive functioning in a group of seniors with a history of falls (Liu-Ambrose, Donaldson, et al., 2008). After six months of the program, participants in the exercise intervention group showed significant improvements in response inhibition scores, as measured by the Stroop, and a reduction in falls over one year, without any corresponding changes in physiological falls risk or mobility. Importantly, both aerobic and resistance training have structural and functional impacts on the brain, although different types of exercise appear to target different neural regions and corresponding executive functions. Thus, once the exact neural circuitry involved in falls is identified, we can determine the type of exercise that may be the most effective for reducing falls risk.  Final Conclusions In conclusion, my dissertation aimed to examine the relationship between altered attentional processing and falls risk in older adults. To that end, my research augments the existing literature to suggest that key  113 factors associated with falls risk include 1. The level of attentional resources available to dedicate to an ongoing task; and 2. The selection of where to allocate these resources. I highlight that the research presented in this dissertation provides preliminary evidence of the relationship between attention and falls, and that my research findings are limited by small and homogeneous samples. Further, the studies presented in this dissertation are correlational, which does not permit conclusions about the causal relationship between these factors. Within this emerging field of research, critical next steps include the identification of plausible mechanisms to explain this relationship and using this information to inform new effective prevention and treatment strategies to be utilized by older adults in our aging population. Specific future studies include a deeper investigation into the hemispheric differences observed between fallers and non-fallers, as well as how cognitive and physical training may reduce falls risk through improvement of executive cognitive functions and corresponding functional plasticity in relevant brain networks.  Advancing into our twilight years is inevitably accompanied by declines in both cognitive and physical function. Understanding the underlying causes of falls and how we can reduce falls risk in the future is of the utmost importance for the health, independence, and well-being of our aging population. Critically, despite what we currently know about the relationship between cognition and falls, one barrier towards the inclusion of cognitive assessments in falls risk screening tests has been the perception of the importance of cognitive factors in falls. Due to the fact that falls are primarily viewed as a physiological problem, cognition is not routinely assessed clinically in falls patients (e.g., Snijders, Verstappen, Munneke, & Bloem, 2007). In a 2001 guideline for falls prevention (Lundebjerg et al., 2001), although cognitive impairment was identified as one of the top risk factors for falls across 16 studies, the guideline failed to include any specific recommendations regarding cognition. Thus, the responsibility rests in researchers to inform clinicians on the important role of cognition in falls, as well as provide effective assessment tools that are both cost and time efficient.   114 References Albert, W. S., Reinitz, M. T., Beusmans, J. M., & Gopal, S. (1999). The role of attention in spatial learning during simulated route navigation. Environment and Planning A, 31, 1459-1472.  Anderson, P. J. (2008). Towards a developmental model of executive function. In V. Anderson, R. Jacobs & P. J. Anderson (Eds.), Executive Functions and the Frontal Lobes (pp. 3-21). New York: Psychology Press. Anstey, K. J., von Sanden, C., & Luszcz, M. A. (2006). An 8-year prospective study of the relationship between cognitive performance and falling in very old adults. Journal of the American Geriatrics Society, 54(8), 1169-1176.  Anstey, K. J., Wood, J., Kerr, G., Caldwell, H., & Lord, S. R. (2009). Different Cognitive Profiles for Single Compared With Recurrent Fallers Without Dementia. Neuropsychology, 23(4), 500-508.  Baddeley, A. (1996). Exploring the central executive. The Quarterly Journal of Experimental Psychology, 49A(1), 5-28.  Baker, L. D., Frank, L. L., Foster-Schubert, K., Green, P. S., Wilkinson, C. W., McTiernan, A., . . . Craft, S. (2010). Effects of aerobic exercise on mild cognitive impairment: A controlled trial. Archives of Neurology, 67(1), 71-79.  Baker, S. C., Rogers, R. D., Owen, A. M., Frith, C. D., Dolan, R. J., Frackowiak, R. S. J., & Robbins, T. W. (1996). Neural systems engaged by planning: A PET study of the Tower of London task. [Article]. Neuropsychologia, 34(6), 515-526.  Ball, K., Berch, D. B., Helmers, K. F., Jobe, J. B., Leveck, M. D., Marsiske, M., . . . Willis, S. L. (2002). Effects of cognitive training interventions with older adults: A randomized controlled trial. The Journal of the American Medical Association, 288(18), 2271-2281.   115 Baloh, R. W., Ying, S. H., & Jacobson, K. M. (2003). A longitudinal study of gait and balance dysfunction in normal older people. Arch Neurol, 60(6), 835-839.  Beauchet, O., Dubost, V., Allali, G., Conthier, R., Hermann, F. R., & Kressig, R. W. (2007). 'Faster counting while walking' as a predictor of falls in older adults. Age and Ageing, 36(4), 418-423.  Bellgrove, M. A., Hester, R., & Garavan, H. (2004). The functional neuroanatomical correlates of response variability: evidence from a response inhibition task. [Article]. Neuropsychologia, 42(14), 1910-1916.  Bhadelia, R. A., Price, L. L., Tedesco, K. L., Scott, T., Qiu, W. Q., Patz, S., . . . Bergethon, P. (2009). Diffusion tensor imaging, white matter lesions, the corpus callosum, and gait in the elderly. Stroke, 40, 3816-3920.  Bloem, B. R., Grimbergen, Y. A., van Dijk, J. G., & Munneke, M. (2006). The "posture second" strategy: A review of wrong priorities in Parkinson's disease. Journal of the Neurological Sciences, 248(1-2), 196-204.  Bloem, B. R., Valkenburg, V. V., Slabbekoorn, M., & van Dijk, J. G. (2001). The multiple tasks test: Strategies in Parkinson's disease. Experimental Brain Research, 127(3-4), 478-486.  Bloem, B. R., Valkenburg, V. V., Slabbekoorn, M., & Willemsen, M. D. (2001). The multiple tasks test: Development and normal strategies. Gait and Posture, 14(3), 191-202.  Boehler, C. N., Appelbaum, L. G., Krebs, R. M., Hopf, J. M., & Woldorff, M. G. (2010). Pinning down response inhibition in the brain - Conjunction analyses of the Stop-signal task. NeuroImage, 52, 1621-1632.  Bootsma-van der Wiel, A., Gussekloo, J., de Craen, A. J. M., van Exel, E., Bloem, B. R., & Westendorp, R. G. J. (2003). Walking and talking as predictors of falls in the general population: The Leiden 85-Plus Study. Journal of the American Geriatrics Society, 51(10), 1466-1471.   116 Botvinick, M. M. (2007). Conflict monitoring and decision making: Reconsiling two perspectives on anterior cingulate function. Cognitive, Affective, and Behavioural Neuroscience, 7(4), 356-366.  Briley, D. P., Wasay, M., Sergent, S., & Thomas, S. (1997). Cerebral white matter changes (leukoaraiosis), stroke, and gait disturbance. J Am Geriatr Soc, 45(12), 1434-1438.  Bublak, P., Redel, P., & Finke, K. (2006). Spatial and non-spatial attention deficits in neurodegenerative diseases: Assessment based on Bundesen's theory of visual attention (TVA). Restorative Neurology and Neuroscience, 24, 287-301.  Buckner, R. L., Andrews-Hanna, J. R., & Schacter, D. L. (2008). The brain's default netowrk: anatomy, function, and relevance to disease. Annual NY Acadamy of Science, 1124, 1-38.  Cabeza, R. (2002). Hemispheric asymmetry reduction in older adults: The HAROLD model. [Review]. Psychology and Aging, 17(1), 85-100.  Callisaya, M. L., Blizzard, L., Schmidt, M. D., McGinley, J. L., & Srikanth, V. K. (2010). Ageing and gait variability - a population-based study of older people. Age and Ageing, 39(2), 191-197.  Carriere, J. S. A., Cheyne, J. A., & Smilek, D. (2008). Everyday attention lapses and memory failures: The affective consequences of mindlessness. [Article]. Consciousness and Cognition, 17(3), 835-847.  Carter, C. S., Braver, T. S., Barch, D. M., Botvinick, M. M., Noll, D., & Cohen, J. D. (1998). Anterior cingulate cortex, error detection, and the online monitoring of performance. Science, 280(5364), 747-749.  Casey, B. J., Trainor, R. J., Orendi, J. L., Schubert, A. B., Nystrom, L. E., Giedd, J. N., . . . Rapoport, J. L. (1997). A developmental functional MRI study of prefrontal activation during performance of a Go-No-Go task. Journal of Cognitive Neuroscience, 9(6), 835-847.  Chapman, G. J., & Hollands, M. A. (2007). Evidence that older adult fallers prioritise the planning of future stepping actions over the accurate execution of ongoing steps during complex locomotor tasks. Gait and Posture, 26(1), 59-67.   117 Charlson, M. E., Pompei, P., Ales, K. L., & MacKenzie, C. R. (1987). A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis, 40(5), 373-383.  Cheyne, J. A., Carriere, J. S. A., & Smilek, D. (2006). Absent-mindedness: Lapses of conscious awareness and everyday cognitive failures. [Article]. Consciousness and Cognition, 15(3), 578-592.  Christoff, K., Gordon, A. M., Smallwood, J., Smith, R., & Schooler, J. W. (2009). Experience sampling during fMRI reveals default network and executive system contributions to mind wandering. [Article]. Proceedings of the National Academy of Sciences of the United States of America, 106(21), 8719-8724.  Colcombe, S. J., Erickson, K. I., Scalf, P. E., Kim, J. S., Prakash, R., McAuley, E., . . . Kramer, A. F. (2006). Aerobic Exercise Training Increases Brain Volume in Aging Humans. J Gerontol A Biol Sci Med Sci, 61(11), 1166-1170.  Colcombe, S. J., Kramer, A. F., Erickson, K. I., Scalf, P., McAuley, E., Cohen, N. J., . . . Elavsky, S. (2004). Cardiovascular fitness, cortical plasticity, and aging. Proceedings of the National Academy of Science, 101(9), 3316-3321.  Corbetta, M., & Shulman, G. L. (2002). Control of goal-directed and stimulus-driven attention in the brain. [Review]. Nature Reviews Neuroscience, 3(3), 201-215.  Craik, F. I. M., & Byrd, M. (1982). Aging and cognitive deficits: The role of attentional resources. In F. I. M. Craik & S. Trehub (Eds.), Aging and cognitive processes (pp. 191-211). New York: Plenum. Curran, T., Hills, A., Patterson, M. B., & Strauss, M. E. (2001). Effects of aging on visuospatial attention: an ERP study. [Article]. Neuropsychologia, 39(3), 288-301.  D'Esposito, M., Detre, J. A., Alsop, D. C., Shin, R. K., Atlas, S., & Grossman, M. (1995). The neural basis of the central executive system of working memory. Nature, 378, 279-281.   118 Dark, V. J., Johnston, W. A., Myles-Worsley, M., & Farah, M. J. (1985). Levels of selection and capacity limits. Journal of Experimental Psychology: General, 114, 472-497.  De Luca, C. R., & Leventer, R. J. (2008). Developmental trajectories of executive functions across the lifespan. In V. Anderson, R. Jacobs & P. J. Anderson (Eds.), Executive Functions and the Frontal Lobes (pp. 24-56). New York: Psychology Press. Delbaere, K., Close, J. C. T., Heim, J., Sachdev, P., Brodaty, H., Slavin, M. J., . . . Lord, S. R. (2010). A multifactorial approach to understanding fall risk in older people. Journal of the American Geriatrics Society, 58(9), 1679-1685.  Delorme, A., & Makeig, S. (2004). EEGLAB: An open source toolbox for analysis of single-trial EEG dynamics. Journal of Neuroscience Methods, 134, 9-21.  Dempster, F. N. (1992). The Rise and Fall of the Inhibitory Mechanism - toward a Unified Theory of Cognitive-Development and Aging. [Review]. Developmental Review, 12(1), 45-75.  Donchin, E., & Coles, M. G. H. (1988). Is the P300 Component a Manifestation of Context Updating. [Article]. Behavioral and Brain Sciences, 11(3), 357-374.  Eimer, M. (1996). ERP modulations indicate the selective processing of visual stimuli as a result of transient and sustained spatial attention. Psychophysiology, 33, 13-21.  Eimer, M. (1998). Mechanisms of visuospatial attention: Evidence from event-related brain potentials. Visual Cognition, 5, 257-286.  Elua, I., Laws, K. R., & Kvavilashvili, L. (2012). From mind-pops to hallucinations? A study of involuntary semantic memories in schizophrenia. [Article]. Psychiatry Research, 196(2-3), 165-170.  Enright, P. L. (2003). The six-minute walk test. Respiratory Care, 48(8), 783-785.  Eriksen, B. A., & Eriksen, C. W. (1974). Effects of noise letters upon the identification of a target letter in a non-search task. Perception and Psychophysics, 16, 143-149.   119 Faulkner, K. A., Redfern, M. S., Cauley, J. A., Landsittel, D. F., Studenski, S. A., Rosano, C., . . . Hlth, A., Body Composition S. t. u. d. (2007). Multitasking: Association between poorer performance and a history of recurrent falls. Journal of the American Geriatrics Society, 55(4), 570-576. doi: Doi 10.1111/J.1532-5415.2007.01147.X Filler, M. S., & Giambra, L. M. (1973). Daydreaming as a Function of Cueing and Task Difficulty. [Article]. Perceptual and Motor Skills, 37(2), 503-509.  Folstein, J. R., & Van Petten, C. (2008). Influence of cognitive control and mismatch on the N2 component of the ERP: A review. Psychophysiology, 45, 152-170.  Folstein, M. F., Folstein, S. E., & McHugh, P. R. (1975). "Mini-mental state." A practical method for grading the cognitive state of patients for the clinician. Journal of Psychiatric Research, 12, 189-198.  Foundation), T. H. G. S. (1998). The economic burden of unintentional injury in Canada.  Franklin, M. S., Smallwood, J., & Schooler, J. W. (2011). Catching the mind in flight: Using behavioral indices to detect mindless reading in real time. [Article]. Psychonomic Bulletin & Review, 18(5), 992-997.  Frederiksen, K. S., Garde, E., Skimminge, A., Barkhof, F., Scheltens, P., van Straaten, E. C. W., . . . Waldemar, G. (2011). Corpus callosum tissue loss and development of motor and global cognitive impairment: The LADIS study. Dementia and Geriatric Cognitive Disorders, 32, 279-286.  Friston, K. J., Ashburner, J., Frith, C. D., Poline, J.-P., & Heather, J. D. (1995). Spatial registration and normalization of images. Huamn Brain Mapp, 2, 165-189.  Friston, K. J., Frith, C. D., Turner, R., & Frackowiak, R. S. (1995). Characterizing evoked hemodynamics with fMRI. Neuroimage, 2(2), 157-165.  Gazzaniga, M. S., Ivry, R., & Mangun, G. R. (1998). Fundamentals of Cognitive Neuroscience: W.W. Norton.  120 Giambra, L. M. (1989). Task-Unrelated Thought Frequency as a Function of Age - a Laboratory Study. [Article]. Psychology and Aging, 4(2), 136-143.  Giambra, L. M. (1993). The Influence of Aging on Spontaneous Shifts of Attention from External Stimuli to the Contents of Consciousness. [Article]. Experimental Gerontology, 28(4-5), 485-492.  Giambra, L. M. (1995). A Laboratory Method for Investigating Influences on Switching Attention to Task-Unrelated Imagery and Thought. [Article]. Consciousness and Cognition, 4(1), 1-21.  Gibson, M. J., Andres, R. O., Issacs, B., Radebaugh, T., & Worm-Petersen, J. (1987). The prevention of falls in later life. A report of the Kellogg International Work Group on the prevention of falls by the elderly. Danish Medical Bulletin, 34(Suppl 4), 1-24.  Greenwood, P. M., Parasuraman, R., & Alexander, G. E. (1997). Controlling the focus of spatial attention during visual search: Effects of advanced aging and Alzheimer disease. Neuropsychology, 11(1), 3-12.  Greicius, M. D., Srivastava, G., Reiss, A. L., & Menon, V. (2004). Default-mode network activity distinguishes Alzheimer's disease from healthy aging: Evidence from functional MRI. [Article]. Proceedings of the National Academy of Sciences of the United States of America, 101(13), 4637-4642.  Grimley-Evans, J. (1990). Fallers, non-fallers, and Poisson. Age and Ageing, 19, 268-269.  Grodsky, A., & Giambra, L. M. (1990-1991). The consistency across vigilance and reading tasks of individual differences in the occurrence of task-unrelated and task-related images and thoughts. Imagination, Cognition, and Personality, 10, 39-52.  Groll, D. L., To, T., Bombardier, C., & al., e. (2005). The development of a comorbidity index with physical function as the outcome. Journal of Clinical Epidemiology, 58(6), 595-602.  Gruberger, M., Ben-Simon, E., Levkovitz, Y., Zangen, A., & Hendler, T. (2011). Towards a neuroscience of mind-wandering. [Review]. Frontiers in Human Neuroscience, 5.   121 Guralnik, J. M., Ferrucci, L., Simonsick, E. M., Salive, M. E., & Wallace, R. B. (1995). Lower-extremity function in persons over the age of 70 years as a predictor of subsequent disability. New England Journal of Medicine, 332(9), 556-562.  Halligan, P. W., Fink, G. R., Marshall, J. C., & Vallar, G. (2003). Spatial cognition: Evidence from visual neglect. Trends in Cognitive Sciences, 7(3), 125-133.  Handy, T. C., & Mangun, G. R. (2000). Attention and spatial selection: Electrophysiological evidence for modulation by perceptual load. Perception and Psychophysics, 62, 175-186.  Handy, T. C., Schaich Borg, J., Turk, D. J., Tipper, C., Grafton, S. T., & Gazzaniga, M. S. (2005). Placing a tool in the spotlight: Spatial attention modulates visuomotor responses in cortex. NeuroImage, 26, 266-276.  Handy, T. C., Soltani, M., & Mangun, G. R. (2001). Perceptual load and visuocortical processing: Event-related potentials reveal sensory-level selection. Psychological Science, 12(3), 213-218.  Hannan, M. T., Gagnon, M. M., Aneja, J., Jones, R. N., Cupples, L. A., Lipsitz, L. A., . . . Kiel, D. P. (2010). Optimizing the Tracking of Falls in Studies of Older Participants: Comparison of Quarterly Telephone Recall With Monthly Falls Calendars in the MOBILIZE Boston Study. [Article]. American Journal of Epidemiology, 171(9), 1031-1036.  Harlow, J. M. (1868). Recovery from the passage of an iron bar through the head. Publications of the Massachusetts Medical Society, 2, 327-347.  Haupt, S., Axmacher, N., Cohen, M. X., Elger, C. E., & Fell, J. (2009). Activation of the caudal anterior cingulate cortex due to task-related interference in an auditory stroop. Human Brain Mapping, 30(9), 3043-3056.  Hausdorff, J. M., Doniger, G. M., Springer, S., Yogev, G., Giladi, N., & Simon, E. S. (2006). A common cognitive profile in elderly fallers and in patients with Parkinson's disease: The prominence of impaired executive function and attention. Experimental Aging Research, 32(4), 411-429.   122 Helton, W. S. (2009). Impulsive responding and the sustained attention to response task. [Article]. Journal of Clinical and Experimental Neuropsychology, 31(1), 39-47.  Holtzer, R., Friedman, R., Lipton, R. B., Katz, M., Xue, X., & Verghese, J. (2007). The relationship between specific cognitive functions and falls in aging. [Article]. Neuropsychology, 21(5), 540-548.  Hornbrook, M. C., Wingfield, D. J., Stevens, V. J., Hollis, J. F., & Greenlick, M. R. (1991). Falls among older persons: Antecedents and consequences. In R. Weindruch, E. Hadley & M. G. Ory (Eds.), Reducing frailty and falls in older persons. (pp. 1060-1125). Springfield: CC Thomas. Hsu, C. L., Nagamatsu, L. S., Davis, J. C., & Liu-Ambrose, T. (2012). Examining the relationship between specific cognitive processes and falls risk in older adults: A systematic review. Osteoporosis International. doi: 10.1007/s00198-012-1992-z Hsu, C. L., Voss, M. W., Handy, T. C., Davis, J. C., Nagamatsu, L. S., Chan, A., . . . Liu-Ambrose, T. (in prep.). Falls in community-dwelling older adults and disruptions of large-scale brain networks.  Huxhold, O., Li, S. C., Schmiedek, F., & Lindenberger, U. (2006). Dual-tasking postural control: Aging and the effects of cognitive demand in conjunction with focus on attention. Brain Research Bulletin, 69(3), 294-305.  Jackson, J. D., & Balota, D. A. (2012). Mind-Wandering in Younger and Older Adults: Converging Evidence From the Sustained Attention to Response Task and Reading for Comprehension. [Article]. Psychology and Aging, 27(1), 106-119.  James, W. (1890). The Principles of Psychology: Dover Publications. Jenkinson, M. (2003). Fast, automated, N-dimensional phase-unwrapping algorithm. Magn Reson Med, 49(1), 193-197.  Jenkinson, M., & Smith, S. (2001). A global optimisation method for robust affine registration of brain images. Med Image Anal, 5(2), 143-156.  Kahneman, D. (1973). Attention and Effort. Englewood Cliffs, New Jersey: Prentice Hall.  123 Kam, J. W. Y., Dao, E., Farley, J., Fitzpatrick, K., Smallwood, J., Schooler, J. W., & Handy, T. C. (2011). Slow Fluctuations in Attentional Control of Sensory Cortex. [Article]. Journal of Cognitive Neuroscience, 23(2), 460-470.  Kane, M. J., Brown, L. H., McVay, J. C., Silvia, P. J., Myin-Germeys, I., & Kwapil, T. R. (2007). For whom the mind wanders, and when - An experience-sampling study of working memory and executive control in daily life. Psychological Science, 18(7), 614-621.  Killingsworth, M. A., & Gilbert, D. T. (2010). A Wandering Mind Is an Unhappy Mind. [Article]. Science, 330(6006), 932-932.  Kirschner, A., Kam, J. W. Y., Handy, T. C., & Ward, L. M. (2012). Differential synchronization in default and task-networks of the human brain. [Article]. Frontiers in Human Neuroscience, 6.  Kok, A. (2001). On the utility of P3 amplitude as a measure of processing capacity. Psychophysiology, 38, 557-577.  Kramer, A. F., Hahn, S., & Gopher, D. (1999). Task coordination and aging: explorations of executive control processes in the task switching paradigm. Acta Psychol (Amst), 101(2-3), 339-378.  Krampe, R. T., Schaefer, S., Lindenberger, U., & Baltes, P. B. (2011). Lifespan changes in multi-tasking: Concurrent walking and memory search in children, young, and older adults. [Article]. Gait & Posture, 33(3), 401-405.  Krienen, F. M., & Buckner, R. L. (2009). Segregated fronto-cerebellar circuits revealed by intrinsic functional connectivity. Cerebral Cortex, 19, 2485-2497.  Kuo, C. K., Lin, L. Y., Yu, Y. H., Wu, K. H., & Kuo, H. K. (2009). Inverse association between insulin resistance and gait speed in nondiabetic older men: Results from the U.S. National Health and Nutrition Examination Survey (NHANES) 1999-2002. BMC Geriatrics, 9(49), 1471-2318.  Lachenbruch, P. A., Reinsch, S., MacRae, P. G., & Tobis, J. S. (1991). Adjusting for recall bias with the proportional hazards model. Methods of Information in Medicine, 30(2), 108-110.   124 Lamb, S. E., Jorstad-Stein, E. C., Hauer, K., & Becker, C. (2005). Development of a common outcome data set for fall injury prevention trials: The prevention of falls network Europe consensus. [Article]. Journal of the American Geriatrics Society, 53(9), 1618-1622.  Lezak, M. D. (1995). Neuropsychological Assessment (3rd ed.). New York: Oxford University Press. Li, K. Z. H., Roudaia, E., Lussier, M., Bherer, L., Leroux, A., & McKinley, P. A. (2010). Benefits of cognitive dual-task training on balance performance in healthy older adults. Journal of Gerontology: Medical Sciences, 65A(12), 1344-1352.  Lindenberger, U., Marsiske, M., & Baltes, P. B. (2000). Memorizing while walking: Increase in dual-task costs from young adulthood to old age. Psychology and Aging, 15(3), 417-436.  Liu-Ambrose, T., Ahamed, Y., Graf, P., Feldman, F., & Robinovitch, S. N. (2008). Older fallers with poor working memory overestimate their postural limits. Archives of Physical Medicine Rehabilitation, 89, 1335-1340.  Liu-Ambrose, T., Donaldson, M. G., Ahamed, Y., Graf, P., Cook, W. L., Close, J., . . . Khan, K. M. (2008). Otago home-based strength and balance retraining improves executive functioning in older fallers: A randomized controlled trial. [Article]. Journal of the American Geriatrics Society, 56(10), 1821-1830.  Liu-Ambrose, T., Katarynych, L. A., Ashe, M. C., Nagamatsu, L. S., & Hsu, C. L. (2009). Dual-Task Gait Performance Among Community-Dwelling Senior Women: The Role of Balance Confidence and Executive Functions. [Article]. Journals of Gerontology Series a-Biological Sciences and Medical Sciences, 64(9), 975-982.  Liu-Ambrose, T., Nagamatsu, L. S., Graf, P., Beattie, B. L., Ashe, M., & Handy, T. C. (2010). Resistance training and executive functions: A 12-month randomized controlled trial. Archives of Internal Medicine, 170, 170-178.   125 Liu-Ambrose, T., Nagamatsu, L. S., Leghari, A., & Handy, T. C. (2008). Does impaired cerebellar function contribute to risk of falls in seniors? A pilot study using functional magnetic resonance imaging. Journal of American Geriatric Society, 56(11), 2153-2155.  Liu-Ambrose, T. Y., Ashe, M. C., Graf, P., Beattie, B. L., & Khan, K. M. (2008). Increased risk of falling in older community-dwelling women with mild cognitive impairment. Phys Ther, 88(12), 1482-1491.  Lord, S., Clark, R., & Webster, I. (1991). Physiological factors associated with falls in an elderly population. Journal of American Geriatrics Society, 39, 1194-1200.  Lord, S., Menz, H. B., & Tiedemann, A. (2003). A physiological profile approach to falls-risk assessment and prevention. Physical Therapy, 83(3), 237-252.  Lord, S., Sherrington, C., Menz, H., & Close, J. (2007). Falls in Older People. Cambridge: Cambridge University Press. Lord, S., Ward, J., Williams, P., & Anstey, K. (1994). Physiological factors associated with falls in older community-dwelling women. Journal of American Geriatrics Society, 42, 1110-1117.  Lord, S. R., & Fitzpatrick, R. C. (2001). Choice stepping reaction time: A composite measure of falls risk in older people. Journals of Gerontology Series a-Biological Sciences and Medical Sciences, 56(10), M627-M632.  Lovden, M., Schaefer, S., Pohlmeyer, A. E., & Lindenberger, U. (2008). Walking variability and working-memory load in aging: A dual-process account relating cognitive control to motor control performance. Journal of Gerontology: Psychological Sciences, 63B(3), 121-128.  Luck, S. J., Heinze, H. J., Mangun, G. R., & Hillyard, S. A. (1990). Visual event-related potentials index focused attention within bilateral stimulus arrays. II. Functional dissociation of P1 and N1 compponents. Electroencephalography and clinical Neurophysiology, 75, 528-542.  Luck, S. J., Hillyard, S. A., Mouloua, M., Woldorff, M. G., Clark, V. P., & Hawkins, H. L. (1994). Effects of spatial cueing on luminance detectability - Psychophysical and electrophysiological evidence for  126 early selection. Journal of Experimental Psychology - Human Perception and Performance, 20, 887-904.  Lundebjerg, N., Rubenstein, L. Z., Kenny, R. A., Koval, K. J., Martin, F. C., Tinetti, M. E., . . . Young, A. (2001). Guideline for the prevention of falls in older persons. [Article]. Journal of the American Geriatrics Society, 49(5), 664-672.  Lundin-Olsson, L., Nyberg, L., & Gustafson, Y. (1997). "Stops walking when talking" as a predictor of falls in elderly people. The Lancet, 349, 617.  Lundin-Olsson, L., Nyberg, L., & Gustafson, Y. (1998). Attention, frailty, and falls: The effect of a manual task on basic mobility. Journal of the American Geriatrics Society, 46(6), 758-761.  Mangun, G. R., & Hillyard, S. A. (1991). Modulations of sensory-evoked brain potentials indicate changes in perceptual processing during visual spatial priming. Journal of Experimental Psychology - Human Perception and Performance, 17(4), 1057-1074.  Mangun, G. R., Hillyard, S. A., & Luck, S. J. (1993). Electrocortical substrates of visual selective attention. Attention and Performance, 14(14), 219-243.  Mangun, G. R., Luck, S. J., Plager, R., Loftus, W., Hillyard, S. A., Handy, T., . . . Gazzaniga, M. S. (1994). Monitoring the visual world: Hemispheric asymmetries and subcortical processes in attention. Journal of Cognitive Neuroscience, 6, 265-273.  Manly, T., Robertson, I. H., Galloway, M., & Hawkins, K. (1999). The absent mind: Further investigations of sustained attention to response. Neuropsychologia, 37, 661-670.  Masdeu, J. C., Wolfson, L., Lantos, G., Tobin, J. N., Grober, E., Whipple, R., & Amerman, P. (1989). Brain white-matter changes in the elderly prone to falling. Arch Neurol, 46(12), 1292-1296.  Mason, M. F., Norton, M. I., Van Horn, J. D., Wegner, D. M., Grafton, S. T., & Macrae, C. N. (2007). Wandering minds: The default network and stimulus-independent thought. [Article]. Science, 315(5810), 393-395.   127 McCarley, J. S., Vais, M. J., Pringle, H., Kramer, A. F., Irwin, D. E., & Strayer, D. L. (2004). Converstation disrupts change detection in complex traffic scenes. Human Factors, 46(3), 424-436.  McGough, E. L., Kelly, V. E., Logsdon, R. G., McCurry, S. M., Cochrane, B. B., Engel, J. M., & Teri, L. (2011). Associations Between Physical Performance and Executive Function in Older Adults With Mild Cognitive Impairment: Gait Speed and the Timed "Up & Go" Test. [Article]. Physical Therapy, 91(8), 1198-1207.  McKiernan, K. A., D'Angelo, B. R., Kaufman, J. N., & Binder, J. R. (2006). Interrupting the "stream of consciousness": An fMRI investigation. [Article]. NeuroImage, 29(4), 1185-1191.  McVay, J. C., & Kane, M. J. (2009). Conducting the Train of Thought: Working Memory Capacity, Goal Neglect, and Mind Wandering in an Executive-Control Task. [Article]. Journal of Experimental Psychology-Learning Memory and Cognition, 35(1), 196-204.  McVay, J. C., Kane, M. J., & Kwapil, T. R. (2009). Tracking the train of thought from the laboratory into everyday life: An experience-sampling study of mind wandering across controlled and ecological contexts. [Article]. Psychonomic Bulletin & Review, 16(5), 857-863.  Menon, V., & Uddin, L. Q. (2010). Saliency, switching, attention and control: A network model of insula function. Brain Structure and Function, 214, 655-667.  Mirelman, A., Maidan, I., Herman, T., Deutsch, J. E., Giladi, N., & Hausdorff, J. M. (2011). Virtual reality for gait training: Can it induce motor learning to enhance complex walking and reduce fall risk in patients with Parkinson's disease? Journal of Gerontology: Medical Sciences, 66A(2), 234-240.  Miyake, A., Friedman, N. P., Emerson, M. J., Witzki, A. H., & Howerter, A. (2000). The unity and diversity of executive functions and their contributions to complex "frontal lobe" tasks: A latent variable analysis. Cognitive Psychology, 41, 49-100.  Morris, J. C., Rubin, E. H., Morris, E. J., & Mandel, S. A. (1987). Senile Dementia of the Alzheimer Type - an Important Risk Factor for Serious Falls. Journals of Gerontology, 42(4), 412-417.   128 Moscovitch, M. (1992). Memory and Working-with-Memory - a Component Process Model Based on Modules and Central Systems. [Article]. Journal of Cognitive Neuroscience, 4(3), 257-267.  Moscufo, N., Guttmann, C. R. G., Meier, D., Csapo, I., Hildenbrand, P. G., Healy, B. C., . . . Wolfson, L. (2011). Brain regional lesion burden and impaired mobility in the elderly. Neurobiology of Aging, 32, 646-654.  Moscufo, N., Wolfson, L., Meier, D., Liguori, M., Hildenbrand, P. G., Wakefield, D., . . . Guttmann, C. R. G. (2012). Mobility decline in the elderly relates to lesion accrual in the splenium of the corpus callosum. Age, 34, 405-414.  Nagamatsu, L. S., Carolan, P., Liu-Ambrose, T., & Handy, T. C. (2009). Are impairments in visual-spatial attention a critical factor for increased falls risk in seniors? An event-related potential study. Neuropsychologia, 47, 2749-2755.  Nagamatsu, L. S., Carolan, P., Liu-Ambrose, T. Y. L., & Handy, T. C. (2011). Age-related changes in the attentional control of visual cortex: A selective problem in the left visual hemifield. [Article]. Neuropsychologia, 49(7), 1670-1678.  Nagamatsu, L. S., Handy, T. C., Hsu, C. L., Voss, M. W., & Liu-Ambrose, T. (2012). Resistance training promotes cognitive and functional brain plasticity in seniors with probable mild cognitive impairment. Archives of Internal Medicine, 172(8), 666-668.  Nagamatsu, L. S., Voss, M., Neider, M. B., Gaspar, J. G., Handy, T. C., Kramer, A. F., & Liu-Ambrose, T. (2011). Increased cognitive load leads to impaired mobility decisions in seniors at risk for falls. Psychology and Aging, 26(2), 253-259.  Nasreddine, Z. S., Phillips, N. A., Bedirian, V., Charbonneau, S., Whitehead, V., Collin, I., . . . Chertkow, H. (2005). The Montreal Cognitive Assessment, MoCA: A brief screening tool for mild cognitive impairment. Journal of the American Geriatrics Society, 53(4), 695-699.  Nations, U. (2002). World Population Ageing: 1950-2050, New York.   129 Navon, D., & Gopher, D. (1979). On the economy of the human-processing system. Psychological Review, 86(3), 214-255.  Neider, M. B., McCarley, J. S., Crowell, J. A., Kaczmarski, H., & Kramer, A. F. (2010). Pedestrians, vehicles, and cell phones. Accident Analysis and Prevention, 42(2), 589-594.  Park, H.-J., Kim, J. J., Lee, S.-K., Seok, J. H., Chun, J., Kim, D. I., & Lee, J. D. (2008). Corpus callosal connection mapping using cortical gray matter parcellation and DT-MRI. Human Brain Mapping, 29, 503-516.  Pellecchia, G. L. (2005). Dual-task training reduces impact of cognitive task on postural sway. Journal of Motor Behavior, 37(3), 239-246.  Persad, C. C., Giordani, B., Chen, H. C., Ashton-Miller, J. A., Alexander, N. B., Wilson, C. S., . . . Schultz, A. B. (1995). Neuropsychological predictors of complex obstacle avoidance in healthy older adults. Journal of Gerontology Series B - Psychological Sciences and Social Sciences, 50(5), 272-277.  Podsiadlo, D., & Richardson, S. (1991). The timed "Up and Go": A test of basic functional mobility for frail elderly persons. Journal of American Geriatric Society, 39(2), 142-148.  Polich, J. (1987). Task difficulty, probability, and inter-stimulus interval as determinants of P300 from auditory stimuli. Electroencephalography and clinical Neurophysiology, 68, 311-320.  Polich, J. (1996). Meta-analysis of P300 normative aging studies. Psychophysiology, 33, 334-353.  Posner, M. I. (1980). Orienting of Attention. Quarterly Journal of Experimental Psychology, 32(FEB), 3-25.  Powell, L., & Myers, A. (1995). The Activities-Specific Confidence (ABC) Scale. Journal of Gerontology, 50A(1), M28-M34.  Raichle, M. E., & Snyder, A. Z. (2007). A default mode of brain function: A brief history of an evolving idea. NeuroImage, 37(4), 1083-1090.   130 Rapport, L. J., Hanks, R. A., Millis, S. R., & Deshpande, S. A. (1998). Executive functioning and predictors of falls in the rehabilitation setting. Archives of Physical Medicine and Rehabilitation, 79(6), 629-633.  Reichle, E. D., Reineberg, A. E., & Schooler, J. W. (2010). Eye Movements During Mindless Reading. [Article]. Psychological Science, 21(9), 1300-1310.  Resnick, S. M., Pham, D. L., Kraut, M. A., Zonderman, A. B., & Davatzikos, C. (2003). Longitudinal magnetic resonance imaging studies of older adults: A shrinking brain. The Journal of Neuroscience, 23(8), 3295-3301.  Reuter-Lorenz, P. A., Kinsbourne, M., & Moscovitch, M. (1990). Hemispheric control of spatial attention. Brain and Cognition, 12(2), 240-266.  Robertson, I. H., Manly, T., Andrade, J., Baddeley, B. T., & Yiend, J. (1997). Oops: Performance correlates of everyday attentional failures in traumatic brain injured and normal subjects. Neuropsychologia, 35, 747-758.  Robertson, M. C., Campbell, A. J., Gardner, M. M., & Devlin, N. (2002). Preventing Injuries in Older People by Preventing Falls: A Meta-Analysis of Individual-Level Data. J Am Geriatr Soc, 50(5), 905-911.  Rosano, C., Aizenstein, H., Brach, J., Longenberger, A., Studenski, S., & Newman, A. B. (2008). Gait measures indicate underlying focal gray matter atrophy in the brain of older adults. Journal of Gerontology: Medical Sciences, 63A(12), 1380-1388.  Rosano, C., Aizenstein, H. J., Studenski, S., & Newman, A. B. (2007). A regions-of-interest volumetric analysis of mobility limitations in community-dwelling older adults. J Gerontol A Biol Sci Med Sci, 62(9), 1048-1055.  Rosano, C., Brach, J., Longstreth Jr, W. T., & Newman, A. B. (2006). Quantitative measures of gait characteristics indicate prevalence of underlying subclinical structural brain abnormalities in high-functioning older adults. Neuroepidemiology, 26(1), 52-60.   131 Rosano, C., Brach, J., Studenski, S., Longstreth, W. T., Jr., & Newman, A. B. (2007). Gait variability is associated with subclinical brain vascular abnormalities in high-functioning older adults. Neuroepidemiology, 29(3-4), 193-200.  Rosano, C., Simonsick, E. M., Harris, T. B., Kritchevsky, S. B., Brach, J., Visser, M., . . . Newman, A. B. (2005). Association between physical and cognitive function in healthy elderly: the health, aging and body composition study. Neuroepidemiology, 24(1-2), 8-14.  Ryberg, C., Rostrup, E., Paulson, O. B., Barkhof, F., Scheltens, P., van Straaten, E. C. W., . . . Jokinen, H. (2011). Corpus callosum atrophy as a predictor of age-related cognitive and motor impairment: A 3-year follow-up of the LADIS study cohort. Journal of the Neurological Sciences, 307, 100-105.  Ryberg, C., Rostrup, E., Stegmann, M. B., Barkhof, F., Scheltens, P., van Straaten, E. C. W., . . . Waldemar, G. (2007). Clinical significance of corpus callosum atrophy in a mixed elderly population. Neurobiology of Aging, 28, 955-963.  Schooler, J. W. (2002). Re-representing consciousness: dissociations between experience and meta-consciousness. Trends in Cognitive Sciences, 6(8), 339-344.  Schooler, J. W., Smallwood, J., Christoff, K., Handy, T. C., Reichle, E. D., & Sayette, M. A. (2011). Meta-awareness, perceptual decoupling and the wandering mind. [Review]. Trends in Cognitive Sciences, 15(7), 319-326.  Seeley, W. W., Menon, V., Schatzberg, A. F., Keller, J., Glover, G. H., Kenna, H., . . . Greicius, M. D. (2007). Dissociable intrinsic connectivity networks for salience processing and executive control. Journal of Neuroscience, 27(9), 2349-2356.  Shaw, G. A., & Giambra, L. (1993). Task-Unrelated Thoughts of College-Students Diagnosed as Hyperactive in Childhood. [Article]. Developmental Neuropsychology, 9(1), 17-30.   132 Sheridan, P. L., & Hausdorff, J. M. (2007). The role of higher-level cognitive function in gait: Executive dysfunction contributes to fall risk in Alzheimer's Disease. Dementia and Geriatric Cognitive Disorders, 24, 125-137.  Shumway-Cook, A., Brauer, S., & Woollacott, M. (2000). Predicting the probability for falls in community-dwelling older adults using the Timed Up and Go test. Physical Therapy, 80(9), 896-903.  Shumway-Cook, A., Woollacott, M., Kerns, K. A., & Baldwin, M. (1997). The effects of two types of cognitive tasks on postural stability in older adults with and without a history of falls. Journal of Gerontology: Medical Sciences, 52A(4), M232-M240.  Silsupadol, P., Lugade, V., Shumway-Cook, A., van Donkelaar, P., Chou, L.-S., Mayr, U., & Woollacott, M. H. (2009). Training-related changes in dual-task walking performance of elderly persons with balance impairment: A double-blind, randomized controlled trial. Gait & Posture, 29, 634-639.  Skelton, D. A., & Todd, C. (2004). What are the main risk factors for falls amongst older people and what are the most effective interventions to prevent these falls? How should interventions to prevent falls be implemented? In W. H. O. H. E. Network (Ed.), World Health Organisation. Copenhagen: World Health Organization. Smallwood, J. (in press). Differentiating process from occurrence in studies of self-generated thought. Distinguishing how from why the mind wanders: A process-occurrence framework for self-generated mental activity. Psychological Bulletin.  Smallwood, J., Beach, E., Schooler, J. W., & Handy, T. C. (2008). Going AWOL in the brain: Mind wandering reduces cortical analysis of external events. [Article]. Journal of Cognitive Neuroscience, 20(3), 458-469.  Smallwood, J., Davies, J. B., Heim, D., Finnigan, F., Sudberry, M., O'Connor, R., & Obonsawin, M. (2004). Subjective experience and the attentional lapse: Task engagement and disengagement during sustained attention. [Article]. Consciousness and Cognition, 13(4), 657-690.   133 Smallwood, J., O'Connor, R. C., Sudbery, M. V., & Obonsawin, M. (2007). Mind-wandering and dysphoria. [Article]. Cognition & Emotion, 21(4), 816-842.  Smallwood, J., & Schooler, J. W. (2006). The restless mind. [Article]. Psychological Bulletin, 132(6), 946-958.  Smith, S. M. (2002). Fast robust automated brain extraction. Hum Brain Mapp, 17(3), 143-155.  Smith, S. M., Jenkinson, M., Woolrich, M. W., Beckmann, C. F., Behrens, T. E. J., Johansen-Berg, H., . . . Matthews, P. M. (2004). Advances in functional and structural MR image analysis and implementation as FSL. NeuroImage, 23(Supplement 1), S208-S219.  Smith, S. M., Zhang, Y., Jenkinson, M., Chen, J., Matthews, P. M., Federico, A., & De Stefano, N. (2002). Accurate, robust, and automated longitudinal and cross-sectional brain change analysis. Neuroimage, 17(1), 479-489.  Snijders, A. H., Verstappen, C. C., Munneke, M., & Bloem, B. R. (2007). Assessing the interplay between cognition and gait in the clinical setting. Journal of Neural Transmission, 114(10), 1315-1321.  Sonuga-Barke, E. J. S., & Castellanos, F. X. (2007). Spontaneous attentional fluctuations in impaired states and pathological conditions: A neurobiological hypothesis. Neuroscience and Biobehavioral Reviews, 31(7), 977-986.  Soumare, A., Elbaz, A., Zhu, Y., Maillard, P., Crivello, F., Tavernier, B., . . . Tzourio, C. (2009). White matter lesions volume and motor performances in the elderly. Ann Neurol, 65(6), 706-715.  Springer, S., Giladi, N., Peretz, C., Yogev, G., Simon, E. S., & Hausdorff, J. M. (2006). Dual-tasking effects on gait variability: The role of aging, falls, and executive function. Movement Disorders, 21(7), 950-957.  Stawarczyk, D., Majerus, S., Maj, M., Van der Linden, M., & D'Argembeau, A. (2011). Mind-wandering: Phenomenology and function as assessed with a novel experience sampling method. [Article]. Acta Psychologica, 136(3), 370-381.   134 Stuss, D. T., & Alexander, M. P. (2000). Executive functions and the frontal lobes: a conceptual view. Psychological Research, 63, 289-298.  Stuss, D. T., Gow, C. A., & Hetherington, C. R. (1992). No Longer Gage: Frontal-Lobe Dysfunction and Emotional Changes. [Review]. Journal of Consulting and Clinical Psychology, 60(3), 349-359.  Stuss, D. T., Murphy, K. J., Binns, M. A., & Alexander, M. P. (2003). Staying on the job: the frontal lobes control individual performance variability. [Article; Proceedings Paper]. Brain, 126, 2363-2380.  Talairach, J., & Tournoux, P. (1988). Co-planar Stereotaxic Atlas of the Human Brain. New York: Thieme Medical Publishers. Thal, D. R., Del Tredici, K., & Braak, H. (2004). Neurodegeneration in normal brain aging and disease. Sci Aging Knowledge Environ, 2004(23), pe26.  Tinetti, M. E., Deleon, C. F. M., Doucette, J. T., & Baker, D. I. (1994). Fear of Falling and Fall-Related Efficacy in Relationship to Functioning among Community-Living Elders. [Article]. Journals of Gerontology, 49(3), M140-M147.  Tinetti, M. E., Speechley, M., & Ginter, S. F. (1988). Risk factors for falls among elderly persons living in the community. New England Journal of Medicine, 319(26), 1701-1707.  Townsend, J., & Courchesne, E. (1994). Parietal damage and narrow "spotlight" spatial attention. Journal of Cognitive Neuroscience, 6(3), 220-232.  Travers, B. G., Klinger, M. R., & Klinger, L. G. (2011). Attention and working memory in ASD. In D. Fein (Ed.), The Neuropsychology of Autism. New York: Oxford University Press. Treisman, A. M. (1960). Contextual cues in selective listening. Quarterly Journal of Experimental Psychology, 12, 242-248.  Treisman, A. M. (1969). Strategies and models of selective attention. Psychological Review, 76(3), 282-299.   135 van Doorn, C., Gruber-Baldini, A. L., Zimmerman, S., Hebel, J. R., Port, C. L., Baumgarten, M., . . . Magaziner, J. (2003). Dementia as a risk factor for falls and fall injuries among nursing home residents. Journal of the American Geriatrics Society, 51(9), 1213-1218.  Verghese, J., Buschke, H., Viola, L., Katz, M., Hall, C., Kuslansky, G., & Lipton, R. (2002). Validity of divided attention tasks in predicting falls in older individuals: A preliminary study. Journal of the American Geriatrics Society, 50(9), 1572-1576.  Vogel, E. K., & Luck, S. J. (2000). The visual N1 component as an index of a discrimination process. Psychophysiology, 37, 190-203.  Washburn, R. A., Smith, K. W., Jette, A. M., & al., e. (1993). The Physical Activity Scale for the Elderly (PASE): Development and evaluation. Journal of Clinical Epidemiology, 46(2), 153-162.  Wollacott, M., & Shumway-Cook, A. (2002). Attention and the control of posture and gait: A review of an emerging area of research. Gait and Posture, 16, 1-14.  Woolley, S. M., Czaja, S. J., & Drury, C. G. (1997). An assessment of falls in elderly men and women. Journal of Gerontology: Medical Sciences, 52A(2), M80-M87.  Wright, R. E. (1981). Aging, Divided Attention, and Processing Capacity. [Article]. Journals of Gerontology, 36(5), 605-614.  Yantis, S., & Johnston, J. (1990). On the locus of visual selection: Evidence from focused attention tasks. Journal of Experimental Psychology: Human Perception and Performance, 16, 135-149.  Yesavage, J. A. (1988). Geriatric Depression Scale. Psychopharmacol Bull, 24(4), 709-711.  Yogev-Seligmann, G., Hausdorff, J. M., & Giladi, N. (2008). The role of executive function and attention in gait. Movement Disorders, 23(3), 329-342.  Zhang, Y., Brady, M., & Smith, S. (2001). Segmentation of brain MR images through a hidden Markov random field model and the expectation maximization algorithm. IEEE Trans. on Medical Imaging, 20(1), 45-57. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0165517/manifest

Comment

Related Items