UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Situated cognition : a lens for exploring behaviour change across contexts DiGiacomo, Alessandra 2020

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

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
24-ubc_2020_november_digiacomo_alessandra.pdf [ 7.38MB ]
Metadata
JSON: 24-1.0394328.json
JSON-LD: 24-1.0394328-ld.json
RDF/XML (Pretty): 24-1.0394328-rdf.xml
RDF/JSON: 24-1.0394328-rdf.json
Turtle: 24-1.0394328-turtle.txt
N-Triples: 24-1.0394328-rdf-ntriples.txt
Original Record: 24-1.0394328-source.json
Full Text
24-1.0394328-fulltext.txt
Citation
24-1.0394328.ris

Full Text

 Situated Cognition: A Lens for Exploring Behaviour Change Across Contexts by Alessandra DiGiacomo  Hons. BSc., University of Toronto, 2011 M.A., University of British Columbia, 2013   A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF Doctor of Philosophy in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Psychology)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  September 2020  © Alessandra DiGiacomo, 2020    ii The following individuals certify that they have read, and recommend to the Faculty of Graduate and Postdoctoral Studies for acceptance, the dissertation entitled:  Situated Cognition: A Lens for Exploring Behaviour Change Across Contexts  Submitted by Alessandra DiGiacomo in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Psychology.  Examining Committee: Dr. Alan Kingstone, Professor, Psychology, UBC Supervisor  Dr. Joelle LeMoult, Professor, Psychology, UBC Supervisory Committee Member  Dr. Luke Clark, Professor, Psychology, UBC Supervisory Committee Member Dr. Frances Chen, Professor, Psychology, UBC University Examiner   Dr. Guy Faulkner, Professor, Kinesiology, UBC University Examiner            iii Abstract As theories of situated cognition have been increasingly borne out in the literature, our current understanding of cognition is that it is dynamic and intimately tied to the situational context of the immediate environment. For instance, in the laboratory, participants have a tendency to attend to the location where a cartoon face is looking (Friesen & Kingstone, 1998; Friesen, Moore, & Kingstone, 2005), but very few individuals engage in that same looking behaviour when walking past a stranger in real life (Gallup et al., 2012a). Despite general agreement with the notion that changes in context often invoke changes to cognition and behaviour, decades of research in experimental psychology have been predicated on the assumption that cognition is supported by processes that remain stable across situations (Kingstone, Smilek, & Eastwood, 2008). In the current work, I take the position that when experiments are planned in a way that considers the embedded and situational nature of cognition and behaviour, one can move towards a richer and more thorough understanding of complex real world contexts. Indeed, the situated cognition framework as well as the related ‘nudge theory’ (i.e., the notion that behaviour can be dramatically influenced by subtle changes in context; Sunstein & Thaler, 2008), form the bedrock of my doctoral work. In Chapter 2, I illustrate how a situated approach to cognitive research results in a simple intervention that has important implications for waste management and environmental policy. In Chapter 3, I establish that subtle variations in context can modify the extent to which pro-social behaviours are displayed. In Chapters 4 and 5, I demonstrate that the presence of mobile phones (i.e., technology as situated cognition) is not always a neutral factor when it comes to performance on cognitive tasks. Across these four data chapters and seven empirical investigations, I demonstrate how consideration of the situated cognition   iv framework carries important methodological implications and has the power to uncover cognitive and behavoural effects that may have otherwise been missed.       v Lay Summary It turns out that we can learn a lot more about how a person thinks and behaves if we don’t just study their own mind but also the ways in which their mind connects with the physical world. For example, one of my experiments showed that when recycling and compost bins are made to be part of a person’s immediate environment, they end up recycling and composting a lot more, which means that less garbage ends up in the landfill. Another showed that completing difficult tasks in an environment that includes the presence of your smartphone sometimes affects how well you can complete that task. Overall, this work shows that it is critical that one consider how a person’s environment (e.g., presence of bins, eye trackers, or phones) affects the way they think and behave.    vi Preface All of the work in the present document was conducted in the Brain and Attention Research lab at the University of British Columbia where Alan Kingstone is the Principal Investigator. I conducted and/or supervised all data collection. I analyzed the data and was the primary writer for all the manuscripts. My supervisor, Alan Kingstone, assisted with project conceptualization and suggested edits to the manuscripts. The experiments reported were approved by the University of British Columbia’s Behavioural Research Ethics Board [Toward a More Natural Approach to Attention Research 1-200: H10-00527].   A version of Chapter 2 has been published. DiGiacomo, A., D.W.-L., Lenkic, P.J., Fraser, B., Zhao, J. & Kingstone, A. (2018). Convenience improves composting and recycling rates in high-density residential buildings. Journal of Environmental Planning and Management, 61:2, 309-331, DOI: 10.1080/09640568.2017.1305332. I was the lead investigator and was primarily responsible for research design, data collection, analysis, and manuscript composition. P. Lenkic and B. Fraser were involved in research design and discussions around methodology. A. Kingstone supervised the work. He and J Zhao edited the final manuscript. A version of Chapter 3 will soon be submitted for publication [DiGiacomo, A., Laidlaw, K., & Kingstone, A (Submitted). The implied social presence of an eye tracker encourages pro-social behaviour.] I was the lead investigator and was primarily responsible for research design, data collection, analysis, and manuscript composition. K. Laidlaw was involved in research design and data collection. A. Kingstone supervised the work and edited the final manuscript. A version of Chapter 4 will soon be submitted for publication [DiGiacomo, A., Will, P., Aeschbach, V., Kushlev, K., Zhao, J., & Kingstone, A (Submitted). Does the presence of a smartphone influence performance on cognitive tasks? It depends.] I was the lead investigator and   vii was primarily responsible for research design, supervising data collection, analysis, and manuscript composition. J. Zhao was involved in research design. P. Will and V. Aeschbach  collected the data. A. Kingstone supervised the work and all authors edited the final manuscript. I was the lead investigator for the experiment described in Chapter 5. I was primarily responsible for research design, supervision of data collection, analysis, and manuscript composition. P. Will collected the data. A. Kingstone supervised the work.   viii Table of Contents Abstract .................................................................................................................................... iii  Lay Summary ............................................................................................................................ v  Preface ...................................................................................................................................... vi  Table of Contents ................................................................................................................... viii  List of Tables .......................................................................................................................... xiii  List of Figures ........................................................................................................................ xiv  Acknowledgements .............................................................................................................. xviii  Dedication .............................................................................................................................. xix  Chapter 1: Introduction ........................................................................................................... 1  1.1   Situated Cognition ....................................................................................................... 1  1.2   Nudge Theory.............................................................................................................. 3  1.3   A Comparison of Frameworks ..................................................................................... 5  1.4   Thesis Construction ..................................................................................................... 7  1.5   Chapter Overview........................................................................................................ 9  1.6   Thesis Overview ........................................................................................................ 10  Chapter 2: Convenience improves composting and recycling rates in high-density residential buildings ................................................................................................................ 12  2.1   Introduction ............................................................................................................... 12  2.2   Experiment 1 ............................................................................................................. 18  2.2.1   Methods ............................................................................................................. 19  2.2.1.1   Buildings and Conditions ............................................................................... 19  2.2.1.2   Materials ........................................................................................................ 22    ix 2.2.1.3   Procedure ....................................................................................................... 24  2.2.2   Results and Discussion ...................................................................................... 24  2.2.2.1   Data Handling ................................................................................................ 24  2.2.2.2   Weight of Food Waste Disposed in Kilograms Per Unit Per Week ................. 24  2.3   Experiment 2 ............................................................................................................. 26  2.3.1   Methods ............................................................................................................. 27  2.3.1.1   Buildings and Conditions ............................................................................... 27  2.3.1.2   Materials ........................................................................................................ 34  2.3.1.3   Procedure ....................................................................................................... 35  2.3.2   Results and Discussion ...................................................................................... 35  2.3.2.1   Data Handling ................................................................................................ 35  2.3.2.2   Weight of Waste Disposed in Kilograms Per Person Per Week ...................... 36  2.3.2.3   Containers Recycled in Kilograms Per Person Per Week ................................ 36  2.3.2.4   Paper Recycled in Kilograms Per Person Per Week ........................................ 37  2.3.2.5   Compost in Kilograms Per Person Per Week .................................................. 38  2.4   General Discussion .................................................................................................... 40  Chapter 3: The implied social presence of an eye tracker encourages pro-social behaviour ................................................................................................................................................. 44  3.1   Introduction ............................................................................................................... 44  3.2   The Present Investigation........................................................................................... 45  3.2.1   Methods ............................................................................................................. 46  3.2.1.1   Participants .................................................................................................... 46  3.2.1.2   Apparatus and Stimuli .................................................................................... 46    x 3.2.1.3   Procedure ....................................................................................................... 47  3.2.2   Results & Discussion ......................................................................................... 48  3.2.2.1   Cheating Frequency ....................................................................................... 48  3.2.2.2   Time Spent Cheating ...................................................................................... 49  3.2.2.3   Time Course of Cheating ............................................................................... 50  3.2.2.4   Willingness to Confess ................................................................................... 52  3.3   General Discussion .................................................................................................... 53  Chapter 4: Does the presence of a smartphone influence performance on cognitive tasks? It depends. ................................................................................................................................... 57  4.1   Introduction ............................................................................................................... 57  4.2   Experiment 1 ............................................................................................................. 65  4.2.1   Methods ............................................................................................................. 66  4.2.1.1   Participants .................................................................................................... 66  4.2.1.2   Materials ........................................................................................................ 66  4.2.1.3   Procedure ....................................................................................................... 66  4.2.2   Results and Discussion ...................................................................................... 67  4.3   Experiment 2 ............................................................................................................. 69  4.3.1   Methods ............................................................................................................. 70  4.3.1.1   Participants .................................................................................................... 70  4.3.1.2   Materials ........................................................................................................ 70  4.3.1.3   Procedure ....................................................................................................... 70  4.3.2   Results and Discussion ...................................................................................... 70  4.4   Experiment 3 ............................................................................................................. 72    xi 4.4.1   Methods ............................................................................................................. 74  4.4.1.1   Participants .................................................................................................... 74  4.4.1.2   A Note on Recruitment .................................................................................. 75  4.4.1.3   Materials ........................................................................................................ 75  4.4.1.4   Procedure ....................................................................................................... 75  4.4.2   Results and Discussion ...................................................................................... 76  4.5   General Discussion .................................................................................................... 78  Chapter 5: How do technology-related variables influence performance on cognitive tasks? An evidence-based response to popular fears ........................................................................ 83  5.1   Introduction ............................................................................................................... 83  5.2   The Present Investigation........................................................................................... 92  5.3   Methods .................................................................................................................... 93  5.3.1   Participants ........................................................................................................ 93  5.3.2   Procedure........................................................................................................... 93  5.3.2.1   Cognitive Tasks ............................................................................................. 93  5.3.2.2   Self-report Measures ...................................................................................... 95  5.4   Results and Discussion .............................................................................................. 97  5.4.1   A Note Regarding Chosen Analyses .................................................................. 97  5.4.2   Correlations for Main Study Variables ............................................................... 98  5.4.3   Attention & Inhibition...................................................................................... 103  5.4.4   Executive Functioning / Cognitive Control ...................................................... 104  5.4.5   Impulsivity ...................................................................................................... 107  5.4.6   Working Memory ............................................................................................ 111    xii 5.5   General Discussion .................................................................................................. 113  Chapter 6: General Discussion ............................................................................................. 117  6.1   Chapter Summaries ................................................................................................. 117  6.2   The Research With Regard to Situated Cognition and Nudge Theory ...................... 120  6.3   Implications ............................................................................................................. 122  6.4   Limitations .............................................................................................................. 125  6.5   Conclusions ............................................................................................................. 126  References ............................................................................................................................. 127  Appendix A ........................................................................................................................... 142     xiii List of Tables Table 2-1 Conditions and buildings in Experiment 1…………………………………………    36  Table 2-2 Conditions and buildings in Experiment 2 ………………………………………...    44  Table 4-1 Summary of findings ………………………………………………………………    75  Table 5-1 Pearson correlations for main study variables …………………..………………...   112  Table 5-2 Descriptive statistics for main study variables …….……………………………....  117  Table 5-3 Summary of findings ……………………………………………………………....  133    xiv List of Figures Figure 1-1. Schematic representation of thesis............................................................................. 8  Figure 2-1 Floor plan of Building A (the “least convenient” condition) ..................................... 21  Figure 2-2 Floor plan of Building B (on the left, the “more convenient” condition) and Building C (on the right, the “most convenient” condition) ...................................................................... 22  Figure 2-3 The main garbage disposal room in Building A (left) and Buildings B & C (right) ... 23  Figure 2-4 Small compost bins located by the elevator at the base of Building B (left) and by the elevator on each floor of Building C (right). .............................................................................. 23  Figure 2-5 Weight of compost in kilograms, per bedroom, per week. Most convenient condition produces significantly more compost than both the more convenient and inconvenient conditions. Error bars reflect ± 1 SEM. ....................................................................................................... 25  Figure 2-6 Floor plan of Residence B, East Tower (convenient condition: hallway drop-off). The colored rectangles in the center represent the garbage, compost, paper recycling, and container recycling bins. Each number with a letter represents a suite. ..................................................... 29  Figure 2-7 Floor plan of Residence B, North Tower and South Tower ...................................... 30  Figure 2-8 Floor plan of suites (above) and basement (below) in Residence A, Tower 1. A number represents a suite. ......................................................................................................... 31  Figure 2-9 Floor plan of suites (above) and basement (below) in Residence A, Tower 6. A number represents a suite. ......................................................................................................... 32  Figure 2-10 Floor plan of suites (above) and basement (below) in Residence A, Tower 4. A number represents a suite. ......................................................................................................... 33  Figure 2-11 The container recycling, paper recycling, and compost bins used by residents in Experiment 2. ........................................................................................................................... 34    xv Figure 2-12 The makeshift four-stream recycling station used in convenient (hallway dropoff) condition. These were positioned on each floor of the tower, just outside each suite, making access to the bins convenient. .................................................................................................... 34  Figure 2-13 Weight of container recycling in kilograms, per person, per week. Hallway drop-off condition is significantly greater than all other conditions. LD (41ft) is significantly greater than the temptation condition, where garbage chutes are open (41ft). (Error bars reflect ± 1 SEM) ... 37  Figure 2-14 Weight of paper recycling in kilograms, per person, per week. Hallway condition is significantly greater than all other conditions. (Error bars reflect ± 1 SEM)............................... 38  Figure 2-15 Weight of compost in kilograms, per person, per week. Hallway condition is......... 39  significantly greater than all other conditions. (Error bars reflect ± 1 SEM)............................... 39  Figure 3-1. Experimental set-up. View from ceiling spy camera. ............................................... 46  Figure 3-2 Participants wearing an eye tracker were significantly less likely to cheat compared to participants who were not wearing an eye tracker (p < .001) ..................................................... 49  Figure 3-3 Participants who cheated spent equivalent amounts of time cheating (p >.05). ......... 50  Figure 3-4 An analysis of the proportion of cheating that occurred pre-habituation (during the first five minutes) or post-habituation (after the first five minutes). There is a trend towards participants in the eye tracker condition cheating more often in the post-habituation period. (p = .09). .......................................................................................................................................... 52  Figure 3-5 Participants in the no eye tracker condition who solved a puzzle while cheating were less likely to confess and more likely to lie about it compared to the participants in the eye tracker condition who cheated (p < .05). ................................................................................... 53  Figure 4-1 The effect of randomly assigned phone location on fluid intelligence, as measured by the RSPM. Participants in the “phone present” condition correctly solved fewer matrices than   xvi participants in the “phone absent” or “RA phone present” condition (p = .01). Error bars represent standard error of the mean. ......................................................................................... 69  Figure 4-2 There was no effect of phone location on the number of correctly solved matrices. Scores across all conditions were comparable (p > .05). Error bars represent standard error of the mean. ........................................................................................................................................ 71  Figure 4-3 As was the case in Experiment 2, there was no effect of phone location on the number of correctly solved matrices. Scores across all conditions were comparable (p > .05). Error bars represent standard error of the mean. ......................................................................................... 77  Figure 4-5. A demonstration of how technology use and phone presence interact to predict the number of correctly solve matrices. ........................................................................................... 78  Figure 5-1. The relationship between technology use (total, social, and gaming) and score on the Internet Addiction Test ........................................................................................................... 100  Figure 5-2. The relationship between technology use (total, social, and gaming) and score on the Global Severity Index. ............................................................................................................ 102  Figure 5-3. The relationship between technology use (total, social, and gaming) and OSPAN score as a function of phone presence. ..................................................................................... 105  Figure 5-4. The relationship between internet addiction and OSPAN score as a function of phone presence. ................................................................................................................................. 106  Figure 5-5. The relationship between internet addiction and OSPAN score as a function of phone presence. ................................................................................................................................. 107  Figure 5-6. The relationship between technology use and adjusted pump count on the BART. 108  Figure 5-7. The relationship between depression symptomatology and adjusted pump count on the BART. .............................................................................................................................. 109    xvii Figure 5-8. The relationship between social technology use and adjusted pump count on the BART. .................................................................................................................................... 110  Figure 5-9. The relationship between gaming technology use and adjusted pump count on the BART. .................................................................................................................................... 110  Figure 5-10. The relationship between technology use and digit span performance. ................ 111  Figure 5-11. The relationship between IAT score and N-back performance. ............................ 112     xviii Acknowledgements I remain entirely convinced that I won the research supervisor jackpot in Alan Kingstone, whose intelligence is only surpassed by his kindness. Thanks Alan, of course, for all of your research related leadership and support. But mostly – thank you for championing me and reminding me who I am in the midst of it all. You are simply incredible, and I am so, so lucky.   On a separate but related note, I am very thankful for the BAR Lab, which has been a source of such joy, learning, and connection for me. Special thanks to Kaitlin Laidlaw, Trish Varao-Sousa, Jill Dosso, and Eleni Goodfellow for their solidarity and deep friendship; to Professor Bischof for feeding me, teaching me how to code, cook, use cue cards, and attempting to help me keep my desk organized; and to Paris Will and Vanessa Aeschbach for their expert help in executing this work.  Joelle LeMoult and Luke Clark: your feedback and expertise have greatly improved this document, and your curiosity, thoughtfulness, and encouragement have been so appreciated. Thank you!  To my family – I am beyond grateful for you, for so many reasons, including the fact that you spent two full days ‘reading’ this entire dissertation. ;)  To Nathan, who had the courage to be married to a grad student for 3 years, and the patience to spend 1 of those years flying back and forth to Calgary 36 times to keep me company: I love you.   xix Dedication           To Mom and Dad.   1 Chapter 1:  Introduction Imagine that you are working from home and have just logged on to a Zoom meeting, where a guest speaker will be giving a talk. You notice that after a few minutes, all of your colleagues have turned off their video, so you follow suit. Realizing that you are suddenly very hungry, you head to the kitchen and begin making a sandwich while you listen to the speaker and periodically glance up at your screen. Then, you bring your computer into the living room and sprawl out on the couch while enjoying your lunch. After all, it’s much more comfortable there. Once you finish eating, you pick up the stack of unread mail on your coffee table and open up the first envelope. Enclosed is a reminder from the government to file your annual tax return. The letter also mentions that 90% of taxpayers in your area have already submitted their payments. Feeling slightly guilty, you sit up straight and make a mental note not to keep leaving your taxes to the last minute. You email your accountant to set up a meeting to discuss your return. At that point, you notice that some of your colleagues are back in view, and that the speaker is nearing the end of her talk. You wipe the crumbs off of your shirt, walk back to your home office, and turn on your video. This example illustrates the two theories that form the basis of the current document: namely, ‘Situated Cognition’ and ‘Nudge Theory’. In this introduction, I will define and explain these two frameworks and describe how they relate to each other. Then, I will orient the reader to the construction of the current dissertation and provide a detailed overview of the work.  1.1   Situated Cognition Fundamentally, rather than viewing cognition as an invariant process that is stable across contexts, the situated cognition framework considers cognitive processes to be intimately coupled with the surrounding environment (Smilek, Birmingham, Cameron, Bischof &   2 Kingstone, 2006). This means, accordingly, that an individual’s motivations, strategies, thoughts, and behaviours will shift from one situation to the next. As illustrated in the above example, a simple modification to the context (i.e. having the Zoom video turned off) created a scenario where certain behaviours were much more likely (e.g. preparing a meal, lying down, opening mail) and others much less so (e.g. paying undivided attention to the speaker).  That changes in context often invoke changes to cognition and behaviour may seem obvious to some, but I believe this notion is worth making explicit. After all, decades of research in experimental psychology has been predicated on the assumption that cognition is supported by processes that remain stable across situations (Kingstone, Smilek, Ristic, Friesen & Eastwood, 2003). This assumption has allowed researchers to assert that methodologies designed to investigate processes in highly contrived laboratory settings are tapping into the same fundamental processes that exist in complex and dynamic natural situations (Kingstone, Smilek, and Eastwood, 2008). Ironically, Broadbent – one of the key early figures in experimental psychology – was clear about the need to ensure that psychological research approximated “real life” in some way: “The necessity for some relevance to real life is a worthwhile intellectual discipline” (Broadbent, 1971, p.4). It seems, however, that the trend toward maximizing experimental control has, over time, led to psychological inquiries that may have limited generalizability beyond the lab. Indeed, there are countless examples of studies that are not replicable once a minor change is introduced to the laboratory environment (Soto-Faraco, Morein-Zamir, & Kingstone, 2005; Bindemann, Burton, & Langton, 2008).  Fortunately, when experiments are planned in a way that considers the embedded and situational nature of cognition and behaviour, we can, as a field, move towards more thoroughly understanding processes of interest. For instance, there has been a long-standing assumption that   3 images of eyes or faces were good proxies for evaluating how individuals look at the faces of others real, living individuals. However, more recent research has shown that there are more differences than similarities regarding the ways that we treat either static images or videos of people compared to real-life people. For example, in real life we tend to glance (or avoid looking entirely) at people we don't know (Laidlaw & Kingstone, 2011; Foulsham, Walker, and Kingstone, 2011), but we will actively stare at those same people when they are shown in photos or in videos.  1.2   Nudge Theory  Nudge Theory, popularized in 2008 with the book ‘Nudge’ by Nobel Prize winner, Richard Thaler, is a cornerstone of behavioural economics. Nudge Theory is predicated on the notion that behaviour can be dramatically influenced, for better or for worse, by subtle changes in context and by seemingly insignificant details (Sunstein & Thaler, 2008). The Zoom meeting scenario illustrated two different types of nudges – the first, a ‘social nudge’ occurred when the other people on the zoom call turned off their videos and one followed suit. The second, a ‘normative message nudge’ occurred when one was prompted to initiate the tax return process after reading that most other people had already completed it (Hallsworth, List, Metcalfe, & Vlaev, 2014). Indeed, this latter example has been shown to improve tax compliance in the UK. A key concept in nudge theory is that of the ‘choice architect’, defined by Thaler and Sunstein (2008) as ‘one with the responsibility for organizing the context in which people make decisions’. A famous example of choice architecture comes from the case of the Amsterdam Schiphol Airport Urinals: there, a cleaning manager noticed that there was a significant amount of ‘spillage’ in the men’s bathrooms. He decided to etch images of flies onto the center of each   4 urinal, thinking that people might be nudged to shift their ‘aim’ towards the fly. Indeed, the fly-in-urinal etchings reduced spillage by 80% (Thaler & Sunstein, 2008).  According to Thaler and Sunstein (2008), the general intention behind nudge theory is best explained by ‘libertarian paternalism’. Libertarian paternalism suggests that on the one hand, individuals should be free to make the decisions they want to make, and that on the other hand, it is not unreasonable for choice architects to attempt to influence decisions that lead to longer, better, or healthier lives. A related concept includes the well-understood notion that human beings tend not to make rational decisions, even when those decisions go against the values that one has identified as important (e.g. weight loss) (Chaiken and Trope, 1999). Thaler and Sunstein (2008) argue that it is legitimate for choice architects to nudge people toward beneficial decisions. They are careful to clarify that in order for something to be classified as a nudge, it must be avoidable and not mandated. They define a nudge as “any aspect of the choice architecture that alters people’s behaviour in a predictable way without forbidding any options or significantly changing their economic incentives” (Thaler & Sunstein, 2008, p.6). While nudge theory was primarily developed with the intention to influence people in a particular direction, Thaler and Sunstein (2008) acknowledge that because ‘everything matters’, many of us are choice architects without even knowing it! Indeed, many nudges are entirely unintentional, and the responsible choice architects may never be aware of the particular context they have created. One example of this has to do with pay structures: organizations may not be aware that by choosing to pay employees biweekly instead of monthly, they are nudging people towards saving more, simply because for two months in the year, three paycheques are received instead of two, and people tend to save those "extra" paycheques.    5 1.3   A Comparison of Frameworks Prior to the 18th century, academic research was mostly a solitary endeavor. Now,  collaboration within most academic disciplines is commonplace (Mirc, Rouzies, & Teerikangas, 2017). Though there have been calls for better organized collaboration across academic disciplines (van Rijnsoever & Hessels, 2011), the practice of interdisciplinary research is not the status quo. In fact, it is generally acknowledged that the process of scientific discovery often occurs in discipline-specific silos. This creates the reality that different disciplines engage with and generate knowledge about the same or similar constructs while using different descriptors. I would like to propose that this phenomenon has occurred for the fields of cognitive science and behavioural economics concerning the frameworks of situated cognition and nudge theory, respectively. Indeed, I submit that these frameworks have both been constructed upon the same foundational notion: context matters. Again, this notion may seem obvious: after all, most people are likely to accept the suggestion that context is significant (e.g., it is OK to laugh at a comedy club but it is not acceptable to laugh at a funeral). However, it may be that this is an idea we accept on the face of it, but ignore in practice.  It seems that to some degree, both situated cognition and nudge theory evolved as a response to the consequence of effectively ‘forgetting’ that context matters. We have already discussed that the situated cognition framework arose out of a need to correct the erroneous assumption that cognitive processes remain stable across situations. In other words, it became necessary to make explicit the fact that cognition does not exist in and of itself – that it moves beyond the boundaries of individual organisms (Robbins & Aydede, 2014); that it is embodied (Shapiro, 2019) and embedded (Clark, 2012); in other words, it is situated. Absent of this consideration, our understanding of cognitive processes falls short. Similarly, nudge theory is   6 heavily grounded in resisting the commonly held assumption that human beings are reliably rational. Though it is of course true that human beings have the capacity to be rational when they employ the conscious and deliberate properties of the Reflective System (for a complete overview of the Reflective vs Automatic Systems, see Kahneman, 2011), many (if not most) behaviours and cognitions are not governed by this system. Instead, they are a product of the Automatic System, which is notorious for not only inducing error, but inducing error unknowingly. Our collective overconfidence in the human ability to make ‘good’ decisions and dismissal of the human tendency to make ‘bad’ decisions is precisely what makes nudge theory so useful: it provides a simple way to circumvent the consequences associated with systematic biases and irrational behaviours.    Despite the substantial overlap that these two frameworks share, they are not identical. I would like to put forth the suggestion that while situated cognition is primarily concerned with description (e.g., people do not treat real eyes and images of eyes the same way), nudge theory is primarily concerned with application (e.g., having a helpful red light on air conditioners makes people more likely to change their filter on time). Equally, situated cognition provides an acknowledgment (e.g. people behave differently when they believe they are alone versus when they believe they are being watched) while nudge theory provides a call to action (e.g., there is too much spillage in the men’s bathroom: how can we reduce it?). While situated cognition is consistently neutral (e.g., foods displayed at eye level are chosen more frequently than foods displayed below or above eye level), nudge theory often takes a position that may invoke morality and agency (e.g., kids in schools should be eating healthier foods, so the veggies should go up front). Relatedly, with nudge theory, as the name would imply, there is often an intentionality or purpose behind the contextual modification. Thaler and Sunstein (2008),   7 however, are clear in specifying that awareness or intentionality are not necessary components of nudge theory, and as will become apparent,  ‘unintentional’ nudges play a significant role in this dissertation. On the other hand, with respect to situated cognition, the concept of intentionality is not relevant, because, as I’ve said, the framework is merely descriptive. In sum, it would be fair to say, then, that nudge theory goes above and beyond situated cognition or that it represents a specific actionable type of situated cognition. Indeed, all nudges are examples of situated cognition, but not all examples of situated cognition are nudges.  1.4   Thesis Construction  This doctoral work was born out of a desire to cultivate and contribute empirically based insights about important issues of our time. I focus on two main issues that present both challenges and opportunities for our global cultural moment: environmental sustainability and the increasingly immersive web. My thesis contains four experimental data chapters (i.e. Chapters 2 – 5 inclusive) that are designed to stand alone as individual manuscripts, in line with the “thesis as a collection of papers” model of doctoral dissertations. In each of these chapters, I deal with and review separate literatures that are either topically unrelated to each other (i.e. sustainable behaviour in Chapter 2 and the presence of mobile phones in Chapter 4) or only tangentially related (i.e., eye tracking technology in Chapter 3 and the cognitive impact of technology in Chapter 5). In each individual paper, I investigate key questions that are relevant to, and timely within, the corresponding literatures (e.g., how does the role of convenience influence sustainable behaviours?), such that the current document includes four self-contained bodies of work. In addition to my curiosity about these unique content areas in and of themselves, I am fundamentally interested in showcasing how the lens of situated cognition and nudge theory can add explanatory value to research areas, primarily by uncovering effects that   8 may otherwise have been missed. Indeed, these two theories are unifying constructs for the thesis as a whole. This document can be thought of as containing five separate stories – the four that are told in each data chapter, and the fifth overarching story that frames the work through the lens of situated cognition and nudge theory (illustrated schematically in Figure 1-1). Indeed, the entirety of my research herein is conducted through the lens of situated cognition. I take the position that this methodology is critical if one is interested in drawing conclusions about cognition and behaviour in complex real world contexts. As shown in Figure 1-1, my doctoral work engages with both elements of nudge theory: on the one hand, in Chapter 2, I set out to nudge participants explicitly; and on the other hand, in Chapters 3 through 5, I set out to discover if there are any unintentional and unexpected nudges that may already be exerting an influence on human cognition and behaviour.   Figure 1-1. Schematic representation of thesis.   9  1.5   Chapter Overview Chapter 2 illustrates how a situated approach to cognitive research results in a simple intervention that has important implications for waste management and environmental policy. This chapter also serves as a ‘proof-of-concept’: critically, without removing the garbage and recycling stations that residents were used to using, I demonstrate that I was able to nudge individuals towards increasing the frequency of their sustainable behaviours.  Chapter 3 reveals that subtle variations in context can modify the extent to which pro-social behaviours are displayed. This finding has implications both for the way that laboratory research is conducted and for consumers and manufacturers of wearable computing. Unlike the planned nudge that was introduced in Chapter 2, Chapter 3 demonstrates that countless researchers (also known as choice architects) may have unwittingly created a context where eye tracking participants act more pro-socially than they typically would. Chapters 4 demonstrates that the simple presence of mobile phones is not a neutral presence when it comes to performance on tasks of fluid intelligence. Unlike the scenario in Chapter 3 where researchers may have been creating unintended ‘nudges’, this work suggests that individuals, depending on their previous level of technology use, may be unwittingly ‘nudging’ themselves towards better or worse performance when they choose to complete cognitively demanding tasks in the presence of their smartphone.  These findings have implications for how we think about the effects of an increasingly immersive web, both with regard to personal factors (e.g., time spent online) and situational factors (e.g., presence of a phone).   10 Chapter 5 extends the work of Chapter 4 such that performance in a number of additional cognitive domains were studied in the context of technology use, mobile phone presence, mental health symptomatology, and internet addiction. This chapter highlights the reality that although some trends are emerging, there remain many unknowns regarding the influence of ubiquitous connectivity on human cognition. This chapter also demonstrates that cognitive domains vary in terms of how ‘nudgeable’ they are.  1.6   Thesis Overview In addition to the specific questions outlined in each individual chapter, my thesis as a whole is built on one foundational question: how do the lenses of situated cognition and nudge theory add explanatory value to research? Throughout the dissertation, I showcase how these frameworks have the power to uncover effects that may have otherwise been missed. For instance, the sustainability literature has tended to focus on education and behaviour modification as agents of change. Generally, it has overlooked suggestions that mere information and knowledge are not enough to incite large-scale change. In contrast, a situated cognition approach reveals that bin placement  - an environmental change rather than educational change - can vastly alter the likelihood of engaging in sustainable behaviour. Similarly, proponents of wearable computing have generally focused on the convenience and entertainment benefits of smart glasses and augmented reality goggles. A situated cognition approach reveals the possibility of an unintended benefit: that wearable computing might make us more pro-social. It also reveals an effect that has gone unnoticed for as long as researchers have used eyetrackers: that participants in eyetracking experiments may be behaving more morally than they do in the “real world”. Lastly, regarding the ubiquity of mobile technologies, concerns have often centered on the notion that device overuse is producing long-term damage to cognitive systems. A   11 situated cognition approach introduces the possibility that being engaged with technological devices may not be the only relevant consideration: the simple presence of a device is consequential too.     12 Chapter 2:  Convenience improves composting and recycling rates in high-density residential buildings 2.1   Introduction Over the past few decades, issues related to composting and recycling have become increasingly prominent. Composting involves the decomposition of organic waste into a material called humus, which can be added to soil to improve quality. Recycling involves reusing materials such as plastics, paper and glass, thereby reducing the amount of waste entering landfills. In Canada, residential waste production is on the rise: Canadians each produced 418kg of waste in 2004 compared to 366kg in 2000. Recycling is also on the rise – Canadians recycled 112kg per person in 2004 compared to 71kg in 2000. Composting also increased from 32 kilograms per person in 2000 to 51 kilograms in 2004 (Stats Can, 2013).  However, despite the increasing recycling and composting rates, the majority of waste produced by Canadians is still sent to landfills, which directly contribute to water, air, and soil pollutions. Previous studies have suggested that landfills are contributing to both environmental and health problems in Canada (Davies & Mazumder, 2003). Handling waste in landfills exposes humans to hazardous emissions from the landfill itself and increases the risk of contracting diseases via inhalation or physical contact of waste (Hossain, Santhanam, Norulaine & Omar, 2011; Giusti, 2009). These landfills affect the environment in numerous ways, including introducing contamination to water (leached heavy metals and synthetic organic compounds), to air (carbon dioxide, methane emission, greenhouse gases, volatile organic compounds released into the atmosphere), and to the soil (heavy metals and synthetic organic compounds leaked into the earth).    13 More generally, organic waste in landfills is one of the main contributors to greenhouse gas emissions as it converts to methane after undergoing anaerobic decomposition (Li, Park & Zhu, 2011; Khalid, Arshad, Anjum, Mahmood & Dawson, 2011). Methane emitted from decomposition in landfills is particularly problematic for global warming, since it can effectively absorb the sun’s heat, warming the atmosphere. In fact, methane is a greenhouse gas that is 21 times more potent than carbon dioxide in terms of its global warming potential. The Environmental Protection Agency has developed the Waste Reduction Model (WARM) to help solid waste planners and organizations track reductions in greenhouse gas emissions from different material management practices (EPA, 2015). According to WARM, landfills have accounted for approximately 16.2% of the total U.S. anthropogenic methane emissions (EPA 2010), and for 20% of Canadian national methane emissions (ECCC, 2014). In Canada alone, 27 megatonnes of carbon dioxide equivalent are generated annually from landfills, of which 20 megatonnes are being emitted into the atmosphere annually (ECCC, 2014). This accounts for 3% of Canada’s total greenhouse gas emissions (ECCC, 2016). While developed countries now have stringent regulations around the management of landfills (Chartier, 2014; Townsend, Powell, Jain, Xu, Tolaymat, & Reinhart, 2015), the increasing financial and environmental cost of landfilling has motivated many municipalities to create recycling and composting programs aimed at reducing the amount of solid waste destined for landfills and to optimize resource recovery (Domina & Koch, 2002; Reschovsky & Stone, 1994). For instance, “zero waste” policies are becoming increasingly popular  (Cole, Osmani, Quddus, Wheatley & Kay, 2014; Hottle, Bilec, Brown & Landis, 2015; Song, Li & Zeng, 2015; Zaman, 2015), and several jurisdictions are even imposing mandatory landfill bans of certain materials (Karak, Bhagat, & Bhattacharyya, 2012; Liu, Ren, Lin & Wang, 2015).    14 Given the urgency of waste problems, it is imperative to identify best practices for increasing recycling and composting adherence, with an eye toward minimizing the adverse environmental consequences of landfilling. Specifically, noting the trend for urban intensification and high-density living (Melia, Parkhurst & Barton, 2011), it seems reasonable and productive for these efforts to be focused on identifying best practices for pro-environmental behavior in high-density buildings like multi-family dwellings (MFDs). Notably, residential waste contributes approximately 40% of landfill contents (Stats Canada, 2013).  There is a general consensus in the literature that residents of MFDs recycle less than residents of single-family dwellings (SFDs: Ando & Gosselin, 2005; De Young, Boerschig, Carney, Dillenbeck, Elster, Horst, Thomson, 1995; Fallde, 2015). While many studies have examined the variables that influence recycling behavior (e.g. the design of waste bins – Duffy & Verges, 2009; information signage – Austin, Hatfield, Grindle & Bailey, 1993; personal attitudes, knowledge, and prior experiences – Tonglet, Phillips & Bates, 2004; atmospherics – Wu, DiGiacomo, Lenkic, Wong & Kingstone, 2016; see also Schultz, Oskamp & Mainieri, 1995 and Osbaldiston & Schott, 2012 for meta-reviews), with respect to the question about residents in MFDs recycling less than those in SFDs, one of the unifying themes is that recycling is less convenient in MFDs (e.g. Ando & Gosselin, 2005; Derksen & Gartrell, 1993; De Young et al., 1995; Fallde, 2015; Margai, 1997). Convenience has been defined as the distance to a recycling station (Ando & Gosselin, 2005), the amount of space available to store recyclables (Westergard, 1996), the ratio of collection bins to households (Stevens, 1999), or certain housing characteristics such as floor level or presence of an elevator (McQuaid and Murdoch, 1996). In all cases, SFDs are considered to be more convenient than MFDs. This difference is critical for the disparity in recycling behaviors between the two dwelling environments.    15 The specific emphasis on convenience in recycling behavior is consistent with the perception that inconvenience is a barrier to recycling. For example, non-recyclers identified personal and household inconveniences (e.g., no curbside pickup, distant drop-off site) as important reasons for not recycling (Vining & Ebreo, 1990). Relatedly, attendees of an ecology conference were offered a discount if they took a sustainability pledge to reduce their resource use at the conference (i.e., bring in a re-usable water bottle). Sixty-two percent of respondents stated that inconvenience was their biggest obstacle when it came to fulfilling the sustainability pledge they signed, although a definition for inconvenience was not provided (Jarchow, Rice, Ritson, & Hargreaves, 2011). Similarly, Wagner (2011) found that 28% of respondents think that increased convenience would prompt them to recycle. Moreover, curbside pickup increased the probability of recycling newspaper and glass by 22% and 37% respectively, compared to only having drop-off centers available (Reschovsky & Stone, 1994).  Although these findings converge on the conclusion that convenience increases recycling, the data is based largely on questionnaires, surveys, or reported behaviour. There is surprisingly little direct evidence that convenience actually increases rates of recycling and composting. There are a few notable exceptions, however, and these studies guided us in the present investigation.  For instance, improved convenience – reduced travel distance and time – increased recycling of paper and pop cans in classrooms by 20% (Ludwig, Gray, & Rowell, 1998), and in offices by 50% (Brothers, Krantz, & McClannahan, 1994). This finding leads us to question whether these convenience interventions are similarly effective in housing environments like MFDs. There are two studies to draw from and the conclusions they suggest are contradictory.    16 According to Yau (2012), the answer is that convenience matters little in MFDs. He obtained data from various property management companies and used an analytical model to compare the total amount of paper, plastic, and metal materials disposed of over the course of one year in 122 Hong Kong high-rise buildings. 35% of the buildings had recycling stations on every floor, and the rest did not. Although exact figures were not reported, Yau (2012) concluded that having recycling bins on every floor did not increase the amount of recyclables collected. There are a number of factors to keep in mind when considering the results of this study. First, details about the 122 buildings were unavailable, meaning that although we know that 35% of the buildings had recycling stations on each floor, it is difficult to make inferences about the relative convenience of each building. For example, it could be that some buildings without recycling stations on each floor could have been perceived as more convenient than those with recycling stations on each floor, due to various factors such as physical size, presence of an elevator, etc. Second, it is unclear how the 122 buildings fared on education about recycling and composting, and whether residents knew about the bins or understood what should go in them. Third, the situational context of this study may explain the null results. As Yau (2012) pointed out, his study coincided with the aftermath of the SARS outbreak, and as the common areas that housed the convenient recycling stations were poorly ventilated, it is possible that residents were deterred from using these bins for the fear of poor environmental hygiene and mismanagement.  In contrast, according to Bernstad (2014) and Lakhan (2016), convenience does play a role when it comes to increasing sustainable behaviors such as composting and recycling. Bernstad (2014) found that composting rates increased by 30% after residents were provided with disposable food waste sorting equipment. In this study, 1632 rental units in Sweden were outfitted with special food waste sorting equipment which consisted of a metal hanger and a   17 holder for paper bags. These holders were attached to the inside of cupboard doors underneath the kitchen sink, enabling residents to simply remove and dispose of the paper bag filled with food waste. Prior to the installation of this equipment, each household composted an average of .66 kilograms of food waste per week. This increased to .99 kilograms of food waste per week just four weeks after the disposable sorting equipment was installed, and results show that the gain was maintained long-term (i.e., up to 26 months later). Lakhan (2016) measured the amount of recycled materials from 12 multi-residential buildings in the Greater Toronto Area over the course of 9 months. Unfortunately, there were no quantitative statistical analyses employed in this study, so it is unclear whether the findings reflect statistically significant changes. Regardless, Lakhan concluded that retrofitting recycling chutes (i.e. adding recycling chutes in an attempt to make recycling more convenient) into the buildings had no impact on recycling rates. Lakhan concluded that the combination of having recycling chutes and placing recycling bins in building lobbies, however, increased recycling rates by 3.8%.  In summary, the literature on convenience suggests the following: 1) there is a widespread belief (most notably among Western scholars) that inconvenience is a barrier to recycling; 2) this belief has only been directly supported by two experimental studies which were conducted in classroom and office settings; and 3) experimental studies from MFDs yield conflicting results: one contradicts the idea that convenience matters, in that having recycling stations located on each floor of a high-rise building did not improve recycling (Yau, 2012), and the other suggests that convenience in the form of special sorting equipment increased proper food waste disposal (Bernstad, 2014). The primary question emerging from the literature is whether composting and recycling rates can be increased by shortening the distance to composting and recycling bins (i.e. by   18 improving convenience)? Since convenience has been shown to change other types of behaviors, such as choosing healthier food options (Hanks, Smith, & Wansink, 2012; Wansink & Hanks, 2013), we hypothesize that placing recycling and composting bins closer to residents’ suites will lead to an increase in recycling and composting rates.  Given the limited experimental evidence and divergent findings, the current study directly manipulated convenience in two randomized field experiments where buildings were randomly assigned to conditions and their composting and recycling behaviors were monitored over time. Our first study manipulated the distance from the compost bins to the suites in a residential building, and measured the amount of compost produced in conditions where the bins were on the same floor of the suites (highly convenient), at the base of the elevator by the building entrance (moderately convenient), or outside of the building (inconvenient). We then replicated and extended this experiment by measuring compost, paper, and container (e.g., glass bottles, jars, plastic bottles) recycling in student residences, introducing other forms of inconvenience such as having different types of bins in different locations, or presenting the residents with the temptation of using one general-purpose garbage chute instead of having to travel to the basement to recycle and compost.  2.2   Experiment 1 In this study we examined composting rates in three multi-family residential buildings by varying convenience. Convenience was manipulated by altering the distance from the entrance (i.e., the door) of each suite to the closest available compost bin. We predicted that increased convenience (i.e., decreasing distance) would increase composting rates.   19 2.2.1   Methods 2.2.1.1   Buildings and Conditions There were three buildings in this study (located in Vancouver, British Columbia) and the buildings were randomly assigned to three conditions. Prior to the study, none of the buildings had a composting program. During the study, each building was fitted with either one or multiple compost bins (described below) and weekly compost pick-up was arranged. This service was provided free of charge to residents by the research assistants. The three buildings were selected for study because they are owned and operated by the same real estate company, are reported to have comparable demographics and physical layouts, and did not have an existing composting program. The buildings were reported to be at full occupancy at the time of data collection during the study (from October to December, 2014). The buildings’ operations manager reported that although precise demographics for the buildings could not be obtained due to privacy restrictions, the residents in all three buildings were equivalent in terms of demographics. These are relatively new buildings (built in 1992 and 1994), and according to the Statistics Canada 2006 census (Census Profile, n.d.), the surrounding area is considered to be an affluent upper class neighborhood, with the highest income bracket listed in the census - between $50,572 and $180,615. The metropolitan area average individual income is $36,123.  In the least convenient condition (Building A), one large (27.5 (width) × 27.5 (depth) × 46 (height) inches) compost bin was placed outside the building in the main garbage disposal area on the ground floor. In the more convenient condition (Building B), one small (11 × 15 × 26 inches) compost bin was placed by the elevator on the ground floor, in addition to the large compost bin in the main garbage disposal area. In the most convenient condition (Building C),   20 the same small (11 × 15 × 26 inches) compost bin was placed by the elevator on each floor, in addition to the large compost bin in the main garbage disposal area. Each building had 4 floors, consisted of both 1- and 2-bedroom units, and contained a centrally located elevator. See Table 1 below for a detailed description of the conditions.  Table 2-1 Conditions and buildings in Experiment 1.   Convenience is defined as a function of distance (i.e. shorter distance is more convenient) Condition Building Description Floors Number of 1-bedroom units Number of 2-bedroom units Least Convenient A One large compost bin located in the main garbage disposal area 4 30 19 More Convenient B One large compost bin located in the main garbage disposal area + one small compost bin by the elevator on the ground floor 4 22 10 Most Convenient C One large compost bin located in the main garbage disposal area + one small compost bin by the elevator on each floor 4 22 10   Since the current study only assessed convenience as a function of distance, we did not manipulate composting practice or equipment within each residential unit. The units themselves were comparable across buildings, with an interior space of between 639 and 671 square feet1 for 1-bedroom apartments and 791 square feet for 2-bedroom apartments. The kitchen spaces were either 85 or 92 sq.ft. for 1-bedroom apartments or 108 sq.ft. for 2-bedroom apartments.                                                 1 3.28 feet = 1 meter   21 All three buildings were located along the same street: Buildings B and C were side by side and Building A was on the same road less than a mile away. Figures 2-1 and 2-2 present the floor plan in each building. As shown in Figure 2-1, Building A contained a main garbage disposal area which was a room with garbage, composting, and recycling bins outside the building on the ground floor. The average Euclidian distance (i.e., not including the vertical distance travelled in the elevator) from a suite door to the nearest compost bin was 130 ft in Building A. This was calculated by adding the average distance from each suite to the elevator on each floor (52 ft), and the distance from the elevator to the main garbage disposal room on the ground floor (78 ft).  Figure 2-1 Floor plan of Building A (the “least convenient” condition)       22 Buildings B and C shared access to the same garbage disposal room, similar to the room in Building A, with one composting bin, and garbage and recycling bins. The room was also located outside the buildings on the ground floor. Figure 2 presents the floor plans in Buildings B  Figure 2-2 Floor plan of Building B (on the left, the “more convenient” condition) and Building C (on the right, the “most convenient” condition)  and C, which were identical. In Building B, the average Euclidian distance from the suite door to the compost bin located by the elevator on the ground floor was 36 ft. The Euclidian distance  from the elevator to the garbage disposal room on the ground floor was 32 ft. In Building C, the average Euclidian distance from the suite door to the compost bin by the elevator on each floor was also 36 ft. The Euclidian distance from the elevator to the garbage disposal room on the ground floor was 137 ft. Despite the long distance from the elevator to the garbage disposal room on the ground floor, the closest compost bins were still 36 ft away from the suite door on average in both Buildings B and C. Thus, they were still more convenient than Building A.  2.2.1.2   Materials The residents in each building all had access to the main garbage disposal room located on the ground floor of each building (see Figure 2-3 for a photo of each room). The relative location of   23 the garbage room is shown in Figures 2-1 and 2-2. Both rooms contained the same four types of disposal bins, namely, garbage, container recycling, paper recycling, and composting. The dimension of the compost bin in the garbage room was 27.5 × 27.5 × 46 inches.   Figure 2-3 The main garbage disposal room in Building A (left) and Buildings B & C (right)   In addition to the compost bins located in the main garbage room, we added a smaller compost bin (11 × 15 × 26 inches) by the elevator on the ground floor in Building B and four small bins on the four floors in Building C by the elevator. These are shown in Figure 2-4.     Figure 2-4 Small compost bins located by the elevator at the base of Building B (left) and by the elevator on each floor of Building C (right).   24 2.2.1.3   Procedure Garbage collection occurred once a week on Thursdays, between late morning and early afternoon. Two research assistants used a DYMO® S250 Digital USB Shipping Scale to weigh all the compost bins in all three buildings in the morning prior to garbage collection. In addition to the Thursday morning weighing, the small compost bins were emptied and weighed on Monday mornings. Each week, therefore, produced twelve data points: the first (Monday) and second (Thursday) weighing of the four small compost bins in Building C and the one small compost bin in Building B, and the weight of the 2 large compost bins from the garbage disposal rooms. This process continued for a period of three and a half months, with the first four weeks in September serving as a pilot period to train the research assistants in weighing the bins and the building staff on how to provide secure access to the assistants every week.  2.2.2   Results and Discussion 2.2.2.1   Data Handling The total amount of composting, in kilograms was calculated per condition per recording period (7 days). An average per bedroom calculation was then made, as there were different numbers of one and two bedroom units in each tower/building. 50% of the contents of the large bin located behind Building’s B and C (see Methods, Experiment 1) were attributed to B and the other 50% to C. This large bin was shared between the two buildings. It was not possible to determine exactly what proportion of waste originated from each building. The following data reflects 10 week-long recording periods (i.e., n = 10 per condition). 2.2.2.2   Weight of Food Waste Disposed in Kilograms Per Unit Per Week A univariate analysis of variance (ANOVA) was conducted, with three conditions of convenience as the independent variable (30 observations total), and amount of compost as the   25 dependent variable. This revealed that the amount of compost produced in each condition differed significantly, F(2,27)=8.23, p<.01, ηp2=.38. Tukey’s multiple comparisons revealed that the most convenient condition produced significantly more compost than both the more convenient elevator (p<.05)2 and inconvenient conditions (p<.05). The results are displayed in Figure 2-5.  Figure 2-5 Weight of compost in kilograms, per bedroom, per week. Most convenient condition produces significantly more compost than both the more convenient and inconvenient conditions. Error bars reflect ± 1 SEM.  As predicted, the most convenient condition resulted in the highest compost diversion rates, suggesting that a relatively short trip to the compost bin increased composting. Interestingly, the more convenient and inconvenient conditions were not significantly different, suggesting that differences in distance only matter up to a certain point. It could be that the overall inconvenience experienced in negotiating this distance, which included waiting for and                                                2  The  same  result  is  obtained  if  we  omit  the  compost  collected  in  the  shared  bin.  That  is,  the  most  convenient  condition  produced  more  compost  that  the  more  convenient  condition  even  when  the  comparison  was  based  only  on  the  small  hallway  bins  (p<.05).    26 travelling on the elevator, was perceived to be equal to the overall inconvenience experienced in travelling to the bin located outside the building, and thus composting rates did not taper off further for longer trips. Even though the only difference between the most and the more convenient conditions was a short elevator ride, residents seem to perceive the latter as quite inconvenient. It is possible that when assessing the convenience of a short elevator ride, residents consider factors that may add effort or time to the trip, such as getting dressed, putting on shoes, putting out candles, and locking the door. These are all actions which may not be required if instead of riding the elevator, the trip consists of walking a few meters down the hall.  2.3   Experiment 2 This experiment extended Experiment 1 in three important ways. First, we broadened our participant sample to university students who live in high-density student residences at the University of British Columbia (UBC), Vancouver, Canada. Second, instead of focusing only on composting, we examined the weight of materials entering three streams of waste: container recycling, paper recycling, and composting. In doing so we aimed to test whether convenience boosts both recycling and composting rates. Third, we introduced other forms of inconvenience, such as placing bins in different locations, or presenting the residents with the temptation of disposing waste in a conveniently located garbage chute instead of traveling down to the basement to recycle and compost.  As such, we made one condition highly convenient as in Experiment 1 by minimizing the physical distance to the bins from each suite, and placing the bins right outside the suites (5 feet away). All the other conditions were made inconvenient by increasing distance (41ft, 97ft, and 163ft), locating the bins in separate locations (i.e., compost bins were located outside the building while the recycling and garbage bins were in the basement), or introducing a temptation   27 condition (i.e., having a garbage chute available on each floor). We predicted that recycling and composting rates would be the highest in the convenient condition, compared to the other inconvenient conditions. 2.3.1   Methods 2.3.1.1   Buildings and Conditions Two student residences located on the UBC campus were used in this experiment. They were composed of multiple separate towers which were randomly assigned to different conditions. Residence A consisted of two high-rise towers and one mid-rise tower. Residence B consisted of three high-rise towers. Thus, there were six towers in the experiment, randomly assigned to four conditions. These particular residences were selected for study primarily due to logistics: they contained enough space to securely store the industrial scale (Brecknell DS100) used for weighing bins; they contained a separate area to store full bins prior to weighing, and the staff were able and willing to accommodate a change to their usual waste management procedures for the duration of the study. The four conditions were: (1) convenient (hallway drop-off), where a disposal station was located just outside each suite in the hallway; (2) inconvenient (longer distance; either 41, 97, or 163 ft), where residents had to travel to the basement of the building to dispose their waste; (3) inconvenient (different bin locations), where residents had to travel to the basement of the building to dispose garbage and recycling, and go outside the building to dispose compost; and (4) inconvenient (temptation), where residents had to travel to the basement of the building to recycle and compost, but could dispose garbage using the chute located outside their suite on each floor. Since the current study assessed convenience as a function of distance, we again did   28 not manipulate composting and recycling practice or equipment within each suite. See Table 2-2 below for a detailed description of the conditions. Table 2-2 Conditions and Buildings in Experiment 2  Condition Residence tower Description Euclidian distance from suite door to bins Floor #s under observation Total number of residents  Convenient: hallway drop-off Residence B, East Tower Compost (C), container recycling (CR), and paper recycling (PR) bins right outside the suite in the hallway 5 ft 14-17 96 Inconvenient: longer distance Residence B, North Tower C, CR & PR in the same basement location 41 ft 1-17 396 Inconvenient: longer distance Residence A, Tower 1 C, CR & PR in the same basement location  97 ft 1-18 340 Inconvenient: longer distance Residence A, Tower 6 C, CR & PR in the same basement location  163 ft 7 277 Inconvenient: different bin locations Residence A, Tower 4 CR & PR in basement, C outside the tower 86 ft 1-18 401 Inconvenient: temptation Residence B, South Tower C, CR & PR in basement + garbage chutes open 41 ft (5 ft from suite door to garbage chute) 1-17 396  Figure 2-6 presents the floor plan of Residence B, East Tower (convenient condition: hallway drop-off). The Euclidian distance from the suite door to the bins in the hallway was 5 ft.    29  Figure 2-6 Floor plan of Residence B, East Tower (convenient condition: hallway drop-off). The colored rectangles in the center represent the garbage, compost, paper recycling, and container recycling bins. Each number with a letter represents a suite.  In Residence B, the North Tower (inconvenient condition: longer distance) and the South Tower (inconvenient condition: temptation) had the same floor plan as that in the East Tower, except without the bins in the hallway. In both towers, the bins were placed in a garbage disposal room in the basement of each tower. Figure 2-7 presents the floor plan of the basement in the North Tower and the South Tower. The floor plan was identical in both towers. The average Euclidian distance from a suite door to the bins in the basement was 41 ft. This was calculated by adding the average distance from each suite door to the elevator on each floor (7 ft), and the distance from the elevator to the garbage disposal room in the basement (34 ft).    30  Figure 2-7 Floor plan of Residence B, North Tower and South Tower  Figure 2-8 presents the suite-level and basement floor plan for Residence A, Tower 1 (inconvenient condition: longer distance). The average Euclidian distance from a suite door to the bins in the basement was 97 ft. This was calculated by adding the average distance from each suite door to the elevator on each floor (76 ft), and the distance from the elevator to the garbage disposal room in the basement (21 ft).   31     Figure 2-8 Floor plan of suites (above) and basement (below) in Residence A, Tower 1. A number represents a suite.  Figure 2-9 presents the suite-level and basement floor plan for Residence A, Tower 6 (inconvenient: longer distance). The average Euclidian distance from a suite door to the bins in the basement was 163 ft. This was calculated by adding the average distance from each suite door to the elevator on each floor (138 ft), and the distance from the elevator to the garbage disposal room in the basement (25 ft).    32       Figure 2-9 Floor plan of suites (above) and basement (below) in Residence A, Tower 6. A number represents a suite.  Figure 2-10 presents the suite-level and basement floor plan for Residence A, Tower 4 (inconvenient: different bin locations). The average Euclidian distance from a suite door to the bins in the basement was 86 ft. This was calculated by adding the average distance from each suite door to the elevator on each floor (78 ft), and the distance from the elevator to the garbage disposal room in the basement (8 ft).   33   Figure 2-10 Floor plan of suites (above) and basement (below) in Residence A, Tower 4. A number represents a suite.  To understand the make-up of the residents in the towers, we administered a questionnaire to the residents which revealed a comparable make up across all towers in both residences. Of the 250 respondents from Residence A, 95% were between the ages of 18 and 24, and of the 315 respondents from Residence B, 90% were between the ages of 18 and 24. Country of origin was similar across the two residences: in both residences, 48% of residents were from North America; 39% of Residence A vs 31% of Residence B students were from Asia; in both residences, 7% of residents are from Europe; and 2% of Residence A and 10% of Residence B were from South America.  In addition, the populations did not differ in terms of the composition of resident program majors, c2(1, N = 379) = 34.86, p = .25.   34 2.3.1.2   Materials Residents in each tower had access to standard container recycling (gray), paper recycling (blue), and compost (green) bins in the basement of each tower, shown in Figure 2-11. Each bin was 22 × 24 × 40 inches.  Figure 2-11 The container recycling, paper recycling, and compost bins used by residents in Experiment 2.  The convenient (hallway drop-off) condition used makeshift recycling stations, shown in Figure 2-12. Each bin was 11 × 20 × 30 inches.   Figure 2-12 The makeshift four-stream recycling station used in convenient (hallway dropoff) condition. These were positioned on each floor of the tower, just outside each suite, making access to the bins convenient.    35 2.3.1.3   Procedure The custodial staff in each residence tagged each full bin (these bins were not sorted for contamination) every week, identifying the type of trash (containers, paper, or compost) and which tower it came from. They then brought the full bins to a common area in the tower, where research assistants weighed the bins. The research assistants used an industrial scale, the Brecknell DS100 (one was stored in both Residence A and B for the duration of the study) and weighed the bins on Tuesdays and Thursdays between 9 and 10am for a period of three months from September to December. September served as a pilot period to train the research assistants in weighing the bins, and communicating with building staff in getting building access every week. The data in the pilot period (September) were not included in the analyses.  2.3.2   Results and Discussion 2.3.2.1   Data Handling The amount of composting, container recycling, and paper recycling in kilograms was calculated per condition in each week. The average weight per person was calculated. This is a slightly different measurement than the one used in Experiment 1, where the dependent variable was average weight per bedroom. This is due to the fact that due to privacy laws, we could not determine the exact number of people living in each suite for Experiment 1, but in the current experiment it is possible to calculate a per person measure. Two of the total 12-week observation period had a missing data point resulting from either a holiday or a change of garbage pickup schedule, so the following data reflects the 10 week-long observation period (i.e., n = 10 for each condition and each stream of recycling and compost).   36 2.3.2.2   Weight of Waste Disposed in Kilograms Per Person Per Week A 6 (condition) x 3 (waste stream) omnibus two-way multivariate ANOVA was conducted and revealed that the amount of waste produced in each condition differed significantly, F(15,144)=5.21, p<.001; Wilk’s Λ=0.302, ηp2=.70. The degree of convenience had a statistically significant effect on the amount of material disposed of in each of the three streams: container recycling (F(5,54)=8.64; p<.001, ηp2=.44), paper recycling (F(5,54)=7.04; p<.001, ηp2=.40), and compost (F(5,54)=8.04; p<.001, ηp2=.43). Tukey’s multiple comparisons were conducted below to determine the exact conditions that were driving the effect. 2.3.2.3   Containers Recycled in Kilograms Per Person Per Week For container recycling, residents in the convenient condition recycled significantly more containers than those in all other inconvenient conditions, (p<.001). None of the inconvenient conditions differed significantly from each other (p>.05).  To test whether having the garbage chute resulted in less container recycling because residents would be seduced to throw them down the chute, a two-tailed independent samples t-test was conducted to compare the 41ft condition with temptation condition which was also 41ft away. The results showed that residents in the 41ft condition recycled more containers than those in the temptation condition (p<.01), despite the fact that the distance to the recycling bin was identical for both conditions. The results are displayed in Figure 2-13.   37  Figure 2-13 Weight of container recycling in kilograms, per person, per week. Hallway drop-off condition is significantly greater than all other conditions. LD (41ft) is significantly greater than the temptation condition, where garbage chutes are open (41ft). (Error bars reflect ± 1 SEM)   2.3.2.4   Paper Recycled in Kilograms Per Person Per Week For paper recycling, Tukey HSD post-hoc tests revealed that the residents in the convenient condition recycled significantly more paper than residents in all other inconvenient conditions (p<.001). There were no significant difference between any of the inconvenient conditions (p>.05). The results are displayed in Figure 2-14.    38  Figure 2-14 Weight of paper recycling in kilograms, per person, per week. Hallway condition is significantly greater than all other conditions. (Error bars reflect ± 1 SEM)   2.3.2.5   Compost in Kilograms Per Person Per Week Similar to container and paper recycling, Tukey HSD post-hoc tests revealed that the residents in the convenient condition composted significantly more than residents in all the other inconvenient conditions (p<.001). None of the inconvenient conditions were significantly different from each other (p>.05). The results are displayed in Figure 2-15.   39  Figure 2-15 Weight of compost in kilograms, per person, per week. Hallway condition is significantly greater than all other conditions. (Error bars reflect ± 1 SEM)  Overall, the results from Experiment 2 showed that the convenient condition consistently resulted in more compost, container, and paper recycling being diverted from the garbage. This supports our hypothesis that improving convenience increases recycling and composting behaviors in high-density residences. Although it is clear that a highly convenient disposal system leads to increased diversion for all streams, the various types of inconvenient disposal systems did not seem to influence diversion in Experiment 2. This implies that there is a threshold for participating in recycling/composting: if residents will recycle and compost only if they perceive the current setup to be convenient enough. The only exception was for container recycling: the temptation condition and the 41ft longer distance (LD) condition both had the same distance from the suite to the bins, with the difference being that the temptation condition had the availability of a garbage chute on each floor. The 41ft LD condition resulted in approximately three times as much container recycling than the temptation condition. However,   40 this result did not hold for paper recycling or compost. The availability of a nearby garbage chute was tempting for residents who were disposing of containers, suggesting that the threshold for recycling avoidance was lower for containers than for paper or compost. This is consistent with Ando and Gosselin’s (2005) discovery that the euclidian distance to recycling stations had a negative effect on container recycling rates but not on paper recycling rates, suggesting that it is easier to recycle paper than containers given that containers may be more cumbersome to carry.  2.4   General Discussion In the current study we tested how convenience influenced recycling and composting rates by manipulating convenience in residential and student residence buildings, and subsequently measuring container recycling, paper recycling, and compost over 10 weeks. We were specifically interested in examining the functional role of convenience when it comes to diverting waste from the landfill in MFDs, and how different forms of convenience may influence recycling and composting.  In Experiment 1 we manipulated convenience by varying the physical distance between the compost bin and the suites. We found a substantial 70% increase in compost when the location of the bins was highly convenient - with one bin on each floor. This was equivalent to diverting 27 kilograms of compostable materials from the landfill per bedroom unit per year.  In Experiment 2, this finding was replicated. The composting rate increased by 139% when the bin was 5 feet away from a suite, compared to when the bins were farther away. Bernstad (2014), as discussed earlier, found a 50% increase in composting after implementing disposable in-suite food waste sorting equipment. Taken together, these results show that composting rates are affected more strongly by decreasing distance to composting bins than by providing residents with in suite equipment. It is important to note, however, that data comparing   41 the contamination rates for both of these strategies is not available. Future studies should examine whether implementing more convenient in-suite equipment and shorter distances to compost bins yields an additive benefit. We also found that frequency of container and paper recycling was 147% and 137% greater, respectively, when the location of the bins was convenient. This increase was equivalent to diverting 14 kilograms of compost, 23 kilograms of containers, and 22 kilograms of paper, from the landfill per person per year. Though more research and analytics are needed, time-course data from this experiment were potentially suggestive of a trend whereby recycling and composting rates increase week-to-week in the convenient condition (i.e. hallway bins), whereas no upward trajectory was observed in any of the inconvenient conditions. In addition to the role that convenience plays in behaviour change, future studies should explore the potential for a relationship between convenience and sustainable-choice related momentum over time, which could additionally influence both thoughts and feelings about sustainable behaviour. Such a dramatic difference in diversion rates is especially impressive given the large scale of the residential complexes. The anonymity that comes from having hundreds of people living in the same building creates an environment that has traditionally been thought to impede recycling behavior (De Young et al., 1995). The current study indicates that a mere change in physical convenience (i.e., decreasing the distance from suite to bin), without any change in social-motivational factors through interventions (e.g. De Young et al., 1995), leads to a profound increase in composting and recycling in large residential complexes.  Further, the increase in composting and recycling behavior is interesting given the participant demographic in our study. Both students, relatively younger with little income, and older and wealthier individuals recycled and composted more when the bins were located in   42 convenient locations. Our effect is thus robust across demographic factors such as age, income, and education, which are generally positively correlated with recycling behavior (Gamba & Oskamp, 1994; Nixon & Saphores, 2009; Owens, Dickerson & Macintosh, 2000). The current results suggest that a simple infrastructure change can have a drastic effect on pro-environmental behavior across demographics such as age and income level. It is, however, important to acknowledge that we were unable to assess pro-environmental behaviour as a function of race, ethnicity, cultural background, and family composition. This is a limitation of the current study as we were not able to ascertain whether convenience is a significant factor for all ethnicities, or whether this effect is culture-specific. We were also unable to assess underlying attitudes towards recycling and composting. Past findings (Bernstad, 2014; Yau, 2012) suggest that underlying attitudes and culture may influence the effectiveness of a convenience intervention.  The current study also raises an intriguing question: is the key ingredient for boosting recycling and composting rates actual convenience (i.e., physical distance to bin) or perceived convenience? Experiment 1 shows that despite the more convenient condition being closer to resident suites than the least convenient condition, recycling and composting rates were comparable. Both conditions required the use of an elevator, which may signal a cost that goes beyond just physical distance (i.e., extra time to turn off the stove, put on shoes, have a conversation with a neighbor, etc). This could explain that despite the most convenient and more convenient conditions being the same distance away from resident suites, the former produced significantly higher composting rates than the latter. Future studies should examine the impact that perceived convenience has on composting rates. It might be possible to boost composting and recycling rates without changing the physical location of garbage bins, by changing the perception of how convenient the bins are to access.    43 In conclusion, to our knowledge we report the first direct evidence that increasing the convenience of bins promotes recycling and composting in MFDs. Using the Waste Reduction Model (WARM) calculator developed by the Environmental Protection Agency (EPA, 2015), we can precisely calculate that over the course of one year, installing bins in hallways in the three residential buildings under study would result in a total emission reduction of 650 MTCO2E (metric tonnes of carbon dioxide). Although conveniently located disposal stations may not promote recycling and composting compliance in all contexts, our experiments provide support for their effectiveness in both university student residences and residential buildings. The current evidence has important implications for waste management, environmental policy makers, urban designers, and architects to work toward making recycling and composting more accessible and convenient for the public, with the ultimate goal of reducing waste destined for landfills and the costs associated with it.      44 Chapter 3:  The implied social presence of an eye tracker encourages pro-social behaviour 3.1   Introduction Social psychologists have long established that the physical presence of other people modifies behaviour. Specifically, the social presence of others tends to encourage conformity to social norms and invite impression management (Guerin, 1986). For example, in public, people are more likely to provide help to others (Riordan et al, 2001) and to alter the way that they express emotions (Buck, Losow, Murphy & Costanzo, 1992). Shoppers have even been shown to purchase more expensive products when they are in the company of other customers compared to when they find themselves shopping alone (Argo, Dahl, & Manchanda, 2005).  Interestingly, these social presence effects may persist even if the ‘presence’ is implied as opposed to embodied. For example, when images of eyes were posted near cash registers, contributions to an honour system box increased (Bateson, Nettle & Roberts, 2006). Similarly, when a desktop background displayed an image of stylized eyes, players in an online dictator game allocated significantly more money to their anonymous competitors (Haley & Fessler, 2005). More recently, when participants were wearing an eye tracker, they were much less willing than their non-eye tracker-wearing counterparts to look at a provocative swimsuit calendar posted nearby (Risko & Kingstone, 2011). This particular study was the first to demonstrate that wearing an eye tracker could induce positive impression management in a manner consistent with other social presence effects.  Eye tracking as a methodology continues to be popularly employed in psychological research. Furthermore, significant investments in wearable computing technologies such as   45 smartglasses, augmented reality (AR) goggles, and mixed reality lenses (i.e. Microsoft HoloLens 2) suggest that these devices will become increasingly accessible to consumers. As such, the potential for the presence of these technologies to play a role in behavior modification has interesting and important implications. For example, can an implied social presence (i.e. an eye tracker) influence behavior beyond simply altering where individuals choose to look? If so, the usual fears associated with increasingly pervasive smart technologies (i.e. privacy issues) may either be tempered by the promise of device-induced pro-social behavior or heightened by the knowledge that simply wearing the device could affect behaviour. At the same time, as is suggested in Nasiopoulos, Risko, Foulsham, and Kingstone (2015), if implied social presence of an eye tracker does influence behaviour, there is a challenge for assumptions of ecological validity. For instance, by presuming that they do not influence behavior, researchers who use eye trackers to investigate human attention and cognition may be unwittingly drawing their conclusions from a contrived rather than natural mode of looking or acting.  3.2   The Present Investigation The goal of the present work is to investigate whether merely wearing an eye tracker can, in addition to modifying looking behaviour, encourage pro-social actions. That is, can an eye tracker act as an implied social presence in a way that induces pro-social behaviour? To explore this question, I assessed participants’ willingness to cheat when they thought their eyes were being tracked (i.e. wearing an eye tracker) and when they did not think their eyes were being tracked (i.e. not wearing an eye tracker). The experimental task involved assembling as many puzzles as possible in the allotted time. To test whether wearing an eye tracker encouraged pro-social behavior, the reverse side of the puzzle cards contained hints which, despite being told to ignore, were available for participants   46 to take advantage of if they chose to. To surreptitiously monitor cheating, a recording camera was hidden in the ceiling.  3.2.1   Methods 3.2.1.1   Participants Ninety-one university students (61 females) from the University of British Columbia participated in the experiment in exchange for course credit or $5 CAD. Following the experiment, all participants were informed about the hidden camera, at which point they had the option to consent to the use of their data. All of the participants provided informed consent. 11 participants were excluded because they guessed the purpose of the study. 3.2.1.2   Apparatus and Stimuli The Applied Science Laboratory MobileEye eye tracker was used for participants in the eye tracker condition. It consisted of a head-mounted system with glasses connected to digital video recorder. Unbeknownst to the participants, this recorder was never turned on. Instead, a ceiling camera recorded the participants as they completed the Tangram puzzles (Figure 3-1). ThinkFun Shape by Shape Creative Pattern Logic Game for Age 8 to Adult was used.   Figure 3-1. Experimental set-up. View from ceiling spy camera.   47 3.2.1.3   Procedure Participants were brought to the testing room. They were randomly assigned to either the eye tracker wearing or no eye tracker group. In the eye tracker condition participants were fitted with the eye tracker and a brief, sham calibration was conducted (participants were unaware that the recording software was not turned on). Participants were then taught how to complete the puzzles. The experimenter demonstrated how to correctly assemble the puzzle pieces into the grid based on the card that was on top of the stack. When the puzzle was complete, the experimenter placed the finished card aside. The next card in the stack was strategically placed “hint side up”, to covertly alert participants to the fact that the back of the cards contained hints. The experimenter proceeded to turn this card over and placed it at the bottom of the stack, saying “Do not start with that one since you saw the answer”. The experimenter then instructed the participants to complete each card one by one and not to look at the hints. She stated that the participants should complete as many cards as they could in twenty minutes and place the solved cards in a pile so that she could count them upon her return. It should be noted that this set-up provided a couple of different ways for participants to cheat: they could peek at the hints to help them solve the cards or they could put un-solved cards directly into the solved pile. The examiner proceeded to leave the room, letting the participant know that she would return in twenty minutes. Following the puzzle task, participants were given a brief questionnaire, which asked them to guess the purpose of the study, state whether or not they wore an eye tracker, describe their experience of the task (e.g., enjoyable, frustrating), and comment on whether or not they cheated during the task.   48 3.2.2   Results & Discussion 3.2.2.1   Cheating Frequency The primary question of interest had to do with understanding whether participants in the no eye tracker condition would be more likely to cheat. Cheating was coded from the video recording by two independent coders. Cheating was defined as either peeking at the solution at the back of the card or by placing unsolved cards or illegitimately solved cards in the solved pile. As expected, given that cheating was quite obvious, inter-rater reliability was perfect: coders agreed on 100% of observations. A chi-square analysis with a single degree of freedom (df=1) was used to determine whether frequency of cheating3 differed between the participants who were wearing an eye tracker and those who were not wearing an eye tracker. In the no eye tracker condition, 63% of the participants cheated on at least one occasion while in the eye tracker condition, only 25.0% of the participants cheated on at least one occasion. The difference was statistically significant, χ2 (1, N = 80) = 11.42, p < .001. This finding is consistent with the idea that an implied social presence -- in the present case, an eye tracker -- can encourage pro-social behaviour.                                                 3  The  results  remain  the  same  whether  cheating  is  defined  simply  as  peeking  at  the  hints  or  by  moving  unsolved  or  illegitimately  solved  cards  into  the  ‘solved’  pile    49  Figure 3-2 Participants wearing an eye tracker were significantly less likely to cheat compared to participants who were not wearing an eye tracker (p < .001)  3.2.2.2   Time Spent Cheating In addition to wondering whether participants in the no eye tracker condition would be more likely to cheat, I wondered whether those same participants would spend more time cheating. As it turns out, the participants in the no eye tracker condition who cheated spent an average of 62 seconds doing so (median = 23.1). That is, 62 seconds were spent peeking at and presumably studying the hints at the back of the card. In the eye tracker condition, the cheating participants spent 78 seconds on average studying the hints (median = 30.0). Figure 3-3 shows that this difference was not statistically significant, t(29) = .37, p > .72, suggesting that wearing an eye tracker did not influence the amount of time participants spent cheating. In other words, once participants made the decision to cheat, the presence or absence of the eye tracker did not affect how long they were willing to spend peeking at the hints.  2562.50102030405060708090100Eye  Tracker No  Eye  TrackerPARTICIPANTS  WHO  CHEAT  AT  LEAST  ONCE  (%)  50  Figure 3-3 Participants who cheated spent equivalent amounts of time cheating (p >.05).   3.2.2.3   Time Course of Cheating Recent work by Nasiopoulos, Risko, Foulsham, and Kingstone (2015) revealed an interesting discovery regarding the nature of the eye-tracker induced prosocial effect. In this case, the authors were focused on the implied social presence of eye trackers as it related strictly to looking behaviour. After replicating Risko and Kingstone’s (2011) initial finding (i.e. that wearing an eye tracker resulted in participants looking less at provocative swim suit calendars), they sought to understand whether this eye-tracker induced social presence effect exerted a strong and sustained influence, or whether the effect was more transient in nature. They discovered that after approximately five minutes of completing an unrelated task while wearing the eye tracker, a habituation effect of sorts occurred. Indeed, participants who were wearing the eye tracker acted just like those who were not wearing the eye tracker in that they freely looked at the provocative stimulus. Essentially, the implied social presence effect was eliminated by having the participants wear the eye tracker for five minutes prior to being exposed to the provocative stimulus. The authors reasoned that rather than a true habituation, where the eye 61.878.30.020.040.060.080.0100.0120.0140.0No  eye  tracker Eye  trackerDURATION	  OF	  CHEATING	  (SECONDS)  51 tracker would essentially be drained of its ‘power’ to exert a social presence effect, the data spoke more accurately to there being a shift of attention away from the eye tracker. Indeed, Nasiopoulos et al (2015) found that by simply shifting participant’s attention back to the eye tracker (e.g., by conducting a second eye tracker recalibration), the initial effect was re-instated, such that participants were once again less willing to look at the provocative stimulus.  These data prompted me to wonder about the time course of cheating in my own study. I was curious about whether the induction of pro-social actions would be transient just like the induction of pro-social looking behaviour was transient. I predicted that, in the eye tracker condition, participants would more often choose to cheat after having worn the eye tracker for five minutes as opposed to before. Of the participants who solved a puzzle while cheating in the no eye tracker condition, 60% began cheating within in the first 5 minutes of the task and 40% began cheating after at least 5 minutes had elapsed. Of the participants who solved a puzzle while cheating in the eye tracker condition, 25% began cheating within the first 5 minutes, while 75% began cheating after at least 5 minutes had elapsed. As shown in Figure 3-4, this difference was trending towards being statistically significant,  χ2 (1, N = 28) = 2.80, p = .09, and suggests the possibility that participants in the eye tracker condition were more likely to cheat once their attention had shifted away from the eye tracker. I think it is likely that this particular analysis is underpowered, given that there were so few “cheaters” in the eye tracker condition to begin with (i.e. n=10 if cheating is defined as simply peeking, and n=6 if cheating is defined as placing a card in the solved pile after it had been peeked at and/or without actually completing the puzzle on the card). Despite not being able to provide conclusive support for the idea that this implied social presence effect is transient, these data are not inconsistent with the notion that eye trackers   52 may only encourage pro-social actions when the wearer is attending to and has not forgotten about the eye tracker.  Figure 3-4 An analysis of the proportion of cheating that occurred pre-habituation (during the first five minutes) or post-habituation (after the first five minutes). There is a trend towards participants in the eye tracker condition cheating more often in the post-habituation period. (p = .09).   3.2.2.4   Willingness to Confess Nasiopoulos et al (2015) easily re-instated the pro-social influence of the eye tracker by bringing attention back to the device while participants were still wearing the eye tracker. I wondered whether that pro-social influence could be re-instated even after the eye tracker was removed by bringing attention back to the decisions participants made while they were wearing the device. To test this, following the task, participants were presented with a questionnaire asking whether they cheated on the task. Result are shown in Figure 3-5. Regarding the individuals who solved a puzzle while cheating, 33% of those in the no eye tracker condition confessed, compared to 83% of those in the eye tracker condition. In other words, 67% of participants in the no eye tracker 602540750102030405060708090100No  Eye  Tracker Eye  TrackerProportion  of  CheatingPre-­‐Habituation Post-­‐Habituation  53 condition lied about cheating, compared to only 17% of participants in the eye tracker condition. This difference was statistically significant, χ2 (1, N = 24) = 4.53, p = .03, and is consistent with the idea that an implied social presence effect can encourage pro-social behavior even after the implied social presence is no longer present. Even when individuals wearing an eye tracker do cheat, which they do less than their non-eye tracker-wearing counterparts, they are more likely to be honest about this if asked (i.e., they are less likely to lie).   Figure 3-5 Participants in the no eye tracker condition who solved a puzzle while cheating were less likely to confess and more likely to lie about it compared to the participants in the eye tracker condition who cheated (p < .05).   3.3   General Discussion The present investigation was prompted by work suggesting that the implied social presence of an eye tracker could influence looking behaviour, such that participants wearing an eye tracker were much less likely to look at a provocative swimsuit calendar (Risko & Kingstone, 2011). I wondered whether, in addition to looking behaviour, the implied social presence of an eye tracker might influence other actions, such as willingness to cheat on a puzzle task. The experiments herein demonstrate that simply wearing an eye tracker encourages 33.383.366.716.70.010.020.030.040.050.060.070.080.090.0100.0No#Eye#Tracker Eye#TrackerProportionConfesssed Lied  54 individuals to behave in a more pro-social manner. Specifically, compared to participants who wore eye trackers, those who did not wear eye trackers were more likely to cheat and were less likely to tell the truth about having cheated. In accordance with the Nasiopoulos et al. (2015) findings that implied social presence effects fade as attention to the tracker is withdrawn, there was a trend towards eye tracker wearing participants cheating later in the task as opposed to earlier. This would suggest the possibility that in the current work the power of the eye tracker to influence behaviour fades over time.  To my knowledge, this was the first demonstration of an implied social presence effect that persists after the implied social presence itself has been removed. Indeed, even after the eye trackers were removed, participants who had been wearing an eye tracker were more likely than those who had never worn an eye tracker to admit to having cheated. In some ways, this result resembles Nasiopoulos’ (2015) finding that by simply shifting a participant’s attention back to the eye tracker  - in the present study, through a post-test question; in Nasiopoulos et al., by performing an eye tracker recalibration - the initial implied social presence effect was re-instated and participants were once again less willing to look at the provocative stimulus. In other ways, this current result extends Nasiopoulos et al.'s (2015) work because it shows that implied social presence effects are not abolished when the implied social presence no longer exists. If we consider social presence effects to be underpinned by attentional mechanisms (Carver & Scheier, 1981), Guerin’s (2010) Feedback Loop theory could represent a relevant model. Guerin posits that when individuals feel observed, attention is in turn deployed inwardly, and the ensuing increase in self-awareness is responsible for the social presence effects that are demonstrated. In the current example, then, it is possible that when the experimenters asked participants whether or not they cheated, the same feedback loop was initialized, resulting in the maintenance of the   55 implied social presence effect. It is important, however, to acknowledge a related limitation of the current study. Though we posit that this implied social presence effect is driving the manner in which participants responded to the post-experiment question of cheating, it is of course possible that there is a simpler explanation. For instance, it could be that participants believe the the eye tracking data provides proof about cheating behaviour, and that it is therefore not in their best interest to lie about it. Certainly, the present study does not uncover or provide evidence for any particular mechanism underlying the results.  The present work also carries methodological and applied implications for the field of wearable computing. The main effect presented herein (i.e. that the presence of an eye tracker induces pro-social behaviour) would have remained undetected without a commitment to studying the situated nature of cognition (Robbins & Aydede, 2009). As I have already discussed, the field of experimental psychology as a whole tends to assume that cognitive processes remain stable across contexts, despite compelling evidence that they do not (Soto-Faraco, Morein-Zamir, & Kingstone, 2005; Bindemann, Burton, & Langton, 2008). As such, the potential for other commonly employed methodologies (e.g., Optitak) to exert social presence effects should not be dismissed. If there are more widespread issues regarding the degree to which our methodologies approximate natural behaviour, we may be making conclusions based on an incomplete (at best) or an incorrect (at worst) sample of behaviour. In contrast to the potentially challenging methodological implications of this work, the applied implications for the field of wearable computing are more positive. Perhaps for good reason, this particular field has typically been mired in privacy-related issues that introduce a degree of risk (perceived or actual) to the consumer. This risk is directly related to the functioning and inherent capability of the wearable computing technology. What has generally not been considered is the possibility that   56 simply wearing wearable computing can have pro-social implications for the wearer. Given the new evidence presented first by Nasiopoulos et al. (2015) and corroborated herein regarding the ease with which the ‘implied social presence power’ of the eye tracker can be re-instated, it is possible that the owners of wearable computing technologies might behave more pro-socially even when they think about or are reminded of their smart goggles or AR glasses.      57 Chapter 4:  Does the presence of a smartphone influence performance on cognitive tasks? It depends.  4.1   Introduction Smartphones are both powerful and ubiquitous: the average smartphone has more computing power than the most sophisticated IBM computer thirty years ago (Levitin, 2014), and more people currently have access to cell phones than to toilets (Worstall, 2013). Indeed, as of 2019, 96% of Americans reported owning a cellphone, and 81% reported owning a smartphone (up from 35% in 2011) (Pew Research Centre, 2019). Similarly, 95% of American teens have access to a mobile phone, with 45% going online “almost constantly”, a figure that has doubled from 2014-2015 data (Anderson & Jiang, 2018).  The emergence of mobile technologies has utterly changed the landscape of social interactions, dating, banking, shopping, fitness tracking, scheduling, and much more. The proliferation of an increasingly immersive web has also, unsurprisingly, birthed an unprecedented reliance on mobile devices.  Some individuals, for example, experience a rise in anxiety when separated from their smartphone (Cheever, Rosen, Carrier, & Chavez, 2014), and 46% of people go as far as to say that they would not be able to live without their phones (Pew Research Centre, 2015). In fact, according to USA data, 91% of people reported never leaving home without their phone, and the average user was found to interact with their device 2617 times per day (Deutsche Telekom, 2012; Winnick, 2016).  The explosion of engagement with the online world presents an exciting opportunity to study how cognitive systems may respond or adapt. That mobile technologies have revolutionized the way we live is undisputed, but a subtler and perhaps more important question   58 remains: how, if at all, is the potential for constant connectivity altering (for better or worse) attentional systems, information processing abilities, memory capacity, and executive control functions?   A relatively small group of researchers have sought to begin answering this question by focusing on the presence of a mobile phone. That is, aside from engaging with a mobile device or being distracted by device related notifications, does the presence of a mobile phone (e.g., simply having it on the table in front of you) affect the status quo? The existing research suggests that the answer is 'yes, it does'. From eliciting poorer evaluations of relationship quality (Przybylski & Weinstein, 2012), to diminishing affection within the parent-child relationship (Rothstein, 2018), to reducing smiles between strangers (Kushlev, Hunter, Proulx, Pressman & Dunn, 2019), and inducing less favourable recollections of conversation satisfaction, the effect of the presence of a smartphone is, in many cases, not neutral.   Prompted by these results, an even smaller group of researchers set out to explore whether the presence of a mobile phone has an impact on cognitive functioning (Thornton et al., 2014; Hartanto & Yang, 2016; Lyngs, 2017; Ward et al., 2017; Aguila, 2019). One popular theory, termed the “brain drain” hypothesis, posits that a finite pool of cognitive resources is taxed by the presence of one’s phone. Specifically, Ward (2017) proposes that the presence of one’s phone causes cognitive resources to be recruited in order to suppress automatic attention to the phone. Because those same resources are now unavailable for other tasks, performance on other cognitive tasks suffers.  While the “brain drain” hypothesis is intuitive, the extant data is mixed and inconclusive. To our knowledge, a total of five studies examining the influence of phone presence on cognitive performance have been published. Across eight experiments, these five studies have assessed   59 five cognitive domains (i.e., fluid intelligence, cognitive flexibility, inhibitory control, working memory capacity, and sustained attention) using eleven different cognitive measures (i.e., simple digit cancellation task, additive digit cancellation task, Stroop tasks, rotation span task, colour-shape switching task, two trail making tests, Go/No-Go task, automated operation span task, stop signal task, and Raven’s standard progressive matrices). The majority of experiments focused on sustained attention, where five out of fifteen (33%) analyses suggested that the simple presence of a phone was associated with a performance decrement. The remainder suggested that there was either no effect of phone presence or a trend toward there being a performance benefit when a phone was present. Two separate studies assessed working memory capacity and found contradictory results, with Ward (2017) presenting evidence for an impairment of performance, and Hartanto (2016) reporting an improvement in performance when one’s phone is present. When assessing inhibitory control, Hartanto (2016) also reported an improvement in performance associated with phone presence. The one study on fluid intelligence suggests that phone presence results in impaired performance. Table 4-1 provides a summary of these findings.       60 Table 4-1. Summary of findings for all studies assessing the impact of mobile phone presence on cognitive performance.   Red boxes indicate no impact; Green boxes indicate impaired performance; Blue boxes indicate enhanced performance. Total column indicates total for impaired performance. SDCT = Simple Digit Cancellation Task; ADCT = Additive Digit Cancellation Task; OS = Automated Operation Span Task; RPM = Raven’s Standard Progressive Matrices; RST = Rotation Span Task; STR= Stroop Task; SWT= Switch Task; G/NG= Go/No-Go Task  Thornton et al (2015) was the first to consider and test whether the presence of a mobile phone could impair performance on cognitive tasks. These authors used a simple digit cancellation task, an additive cancellation task, and two trail making tasks to measure sustained attention. While explaining instructions, the experimenter placed her cellphone beside one participant and a notebook beside the other participant. Information regarding the status of the phone (i.e. whether it was turned on or off) was unavailable. Results showed that for the more demanding digit   Thornton  (2014) Hartanto (2016) Lyngs (2017) Ward (2017) Aguila (2019) Total E1 E2 E1 E2 E1 E1 E2 E1    Phone Parameters Ownership No Yes Yes Yes Yes Yes Yes Yes  Placement  On desk  Unknown Face up on desk Face down on desk Unknown Power Unknown On On On Unknown Notifications Unknown Silent Unknown Silent Unknown       Cognitive Domain Affected Sustained Attention SDCT SDCT  SDCT   SDCT 33% ADCT ADCT  ADCT   ADCT TMT-A TMT-A     TMT-A TMT-B TMT-B     TMT-B      G/NG  Working Memory Capacity    RST     67%      OS OS  Cognitive Flexibility   SWT      0% Inhibitory Control    STR     0% Fluid Intelligence     RPM   100%   61 cancellation task and the more difficult of the two trail making tasks, participants who were faced with the experimenter’s cellphone scored lower than those who were faced with the experimenter’s notebook. These are intriguing results because they suggest that the presence of another person’s cell phone is sufficient to produce a decrease in cognitive performance. Aguila (2019) partially replicated these results: using Thornton et al.’s (2015) procedure (with the exception that the participant’s own phone was used instead of the RA’s phone), she found that phone presence resulted in a performance decrement during the more demanding digit cancellation task. However, Aguila (2019) failed to find any effect of phone presence on performance for either of the trail making tasks. In fact, those results trended towards a boost in performance for participants who were in the presence of a phone. Thornton et al (2015) conducted a follow up experiment, where the same tasks (i.e. digit cancellation and trail making tests) were administered in a classroom and the participants used their own phones. The results of the first study were replicated, with students who had their phones on the desk scoring lower than those who did not. Lyngs (2017) noted methodological weaknesses in the Thornton (2015) study, and sought to replicate those results using an improved experimental design. In Thornton’s first study, experimenters did not know where the participant’s own phone was and whether or not they were receiving notifications. Lyngs (2017) used a cover story about wanting to see how people take pictures of different objects with their phones, and had participants take photos of various objects before starting with the task under investigation. This allowed the experimenter to move the participant phone to the same location on the desk prior to starting the cancellation and trail making tests. Participant’s phones were positioned face up and left powered on during the experiment. Using this refined method, Thornton’s (2015) results were not replicated. There   62 were no differences between the cell phone and control group on any of the tasks. Lyngs (2017) did, however, find that participants who scored high on the cell phone usage questionnaire reported that the task was more fun and exciting when they completed it with their phone present. Similarly, when participants who scored high on possession attachment completed the task, they perceived it to be more effortless when their phone was present.   In another study investigating the cognitive impact of phone presence, Ward (2017) used the Go/No-Go task to measure sustained attention (Bezdjian et al. 2009), the Automated Operation Span task to measure working memory capacity (OSpan; Unsworth et al. 2005), and a 10-item subset of Raven’s Standard Progressive Matrices to measure fluid intelligence (RSPM; Raven, Raven, and Court 1998). Participants were randomly assigned to three conditions, where phones were either in another room (low salience), in their bag (medium salience), or on the desk (high salience). Participants in the ‘desk’ condition were instructed to turn their phones on silent (both ring and vibrate) and to keep them face down on the desk. The authors found no differences in performance on the sustained attention task, suggesting that phone presence did not interfere with participants’ ability to attend. For both the OSpan and the RSPM, on the other hand, Ward (2017) found that phone salience was associated with performance. That is, participants whose phones were located in another room performed better than those whose phones were located on the desk. Overall, data from this study suggest that phone presence may adversely influence working memory capacity and fluid intelligence, but that it does not seem to influence sustained attention. Interestingly, Hartanto and Yang (2016) performed a very similar experiment and came to the exact opposite conclusion: that phone presence improved working memory capacity. These authors also reported evidence that phone presence improves inhibitory control when measured by a version of the Stroop task. Hartanto and Yang (2016) did not   63 specify the placement of the phones in their study, but wrote that they were turned on silent mode.                                                                                                                                                     A review of literature that explores the impact of phone presence on cognitive performance indicates that the available data is mixed, inconclusive, and sometimes even contradictory. The inconsistencies associated with the effect of phone presence exist for every cognitive domain that has been studied more than one time: for example, working memory capacity sometimes improves and sometimes worsens; sustained attention sometimes worsens and is sometimes unaffected. The inconsistencies also persist both across and within the specific tools used to measure various cognitive domains. For example, regardless of whether sustained attention is measured using the additive digit cancellation task or the trail making test, performance sometimes suffers and is sometimes unaffected. In fact, even when the identical measure is employed by a different researcher, previous results are often not replicated. To be sure, further research is needed to clarify if, when, and how phone presence affects performance on cognitive tasks. The obvious question thus becomes: why might we be seeing such inconsistent data?  At first glance, the notion of phone presence seems to be simple enough to examine. But, is it possible that the issue at hand is not merely about presence itself? It is true that along with ‘presence’ come various other factors. For instance, a present phone could be ringing (or vibrating, or lighting up), obscured by another item (or in full view), out of reach (or in reach), silent but capable of providing a notification due to being turned on (or silent and incapable of providing a notification due to being turned off), self-owned (or owned by someone else), or face-up (or face-down). Could it be that in addition to the physical presence of a phone, there   64 should be consideration for various characteristics of the phone itself?  To our knowledge, there is only one study that explicitly considers manipulating various characteristics of a present phone in the context of investigating performance on cognitive tasks. Johannes’ (2019) study assessed response inhibition in three groups: no-visibility-no-notifications, visibility-with-notifications, and visibility-without-notifications. Visibility meant that participant’s phones were on the desk in front of them, while No-Visibility meant that while the participant’s phones were still ‘present’, they were placed in their jacket pocket next to them. Participants were never allowed to interact with their phones, and notifications consisted of vibrations only. Johannes (2019) reported no differences in performance across the groups, suggesting that response inhibition is not influenced by phone visibility or receiving notifications. However, and crucially, because this study did not utilize a baseline ‘phone absent’ condition, there is no way of knowing whether performance was affected similarly in all cases, or whether there was no effect of phone presence in any condition.   Across the five studies reviewed above, there was likely to be considerable variability regarding several different phone parameters. I use the word ‘likely’ because it is not possible to know for certain. For example, four of the five studies did not report on at least one of the following parameters: ownership (self-owned or other-owned), placement (face-up or face-down), power (on or off), and notifications (loud, vibrate, or silent).  Two of the five studies did not report on any of the parameters. So, although there are certainly cases where phone presence adversely affects performance, specific phone parameters that might end up being associated with this effect are unclear. Moreover, the prevailing “brain drain” hypothesis does not necessarily provide guidance as to how phone parameters might play a role on cognitive   65 performance. For instance, should it matter whether a phone is self-owned or not? And should it matter whether the phone is actually on? In sum, the broad aim of the present investigation is to understand why there are so few replicable findings regarding the impact of phone presence on cognitive performance. Experiments 1 and 2 explore whether a careful assessment of phone parameters might provide some clarity. Experiment 3 explores whether the amount of time an individual spends on their phone (outside of the task itself) might also provide clarity.  Finally, because there has been no attempt to replicate the effects of phone presence on fluid intelligence, the present investigation focuses on that as the cognitive domain of interest (Ward et al., 2017).  4.2   Experiment 1 The purpose of Experiment 1 is two-fold. One aim is to replicate the findings of Ward (2017), where performance on tasks of fluid intelligence were found to be impaired by the presence of one’s mobile phone. The other aim is to extend this finding by exploring whether phone ownership is a relevant consideration for understanding the influence of phone presence on fluid intelligence. Accordingly, participants were randomly assigned to one of three groups:  participant's phone present, participant's phone absent, and, to test the specificity of any phone effect, a third condition had the experimenter's phone present.  In all conditions the participants were instructed to complete as many matrices from the Raven’s Standard Progressive Matrices (RSPM; Raven, Raven, and Court, 1998) as possible in 20 minutes.    66 4.2.1   Methods 4.2.1.1   Participants 77 undergraduates from the University of British Columbia (79% female; Mage= 20.1 years; SDage=1.92 years) participated in the experiment in exchange for course credit or monetary reimbursement.  One participant was excluded because he misunderstood the task instructions. Participants were randomly assigned to either the phone present condition (n=27), the phone absent condition (n=25), or the experimenter's phone present condition (n=24). A reverse power analysis, based on Thornton (2014), revealed that the current study was powered at .71. 4.2.1.2   Materials Participants were provided with a 25-item subset of the Raven’s Standard Progressive Matrices (RSPM; Raven, Raven, and Court 1998). The RSPM is a widely employed measure of nonverbal fluid intelligence (Gf), designed to assess one’s capacity to solve novel problems independent of any pre-existing skill or knowledge. On each trial, participants were presented with an incomplete pattern matrix and tasked with selecting the element that best completed the pattern. Participants were also provided with an additional sheet and pen to record their answers.  4.2.1.3   Procedure When participants arrived at the laboratory, they first read through and signed the consent form. Each participant was then randomly assigned to one of three conditions: phone present, phone absent, or experimenter’s phone present. In the phone present condition, participants were asked to put away all of their belongings except for their valuables (in 100% of cases, this included a phone and wallet/purse), which they were instructed to keep with them during the study. So as to not draw attention to the phones, there were no explicit instructions given regarding placement of the phone or expectations around putting phones on silent. None of the participants turned off   67 their phones while in the presence of the experimenter. All participant’s phones were presumed to be turned on and presumed to belong to the participants themselves. In the phone absent and experimenter phone present condition, participants were asked to leave all of their belongings just outside the testing room. They were told that their belongings would be secure. Once participants were in the testing room, they were verbally given instructions on how to complete the RSPM. Participants were told to select the option they believed fit best to complete the matrix. The experimenter verified that each participant could complete the sample item Participants were given 20 minutes to work through the package. They were told to complete as many as they could within this time frame, and the experimenter notified them when the time was up. Participants were then debriefed and provided with course credit or monetary reimbursement. During the experiment, participants were recorded to monitor phone usage during the task. They were duly notified during debriefing, and given the opportunity to have the recording deleted.  4.2.2   Results and Discussion A univariate ANOVA revealed that the number of correctly solved matrices differed significantly as a function of phone condition, F(2,73)=4.79, p=.01, ηp2=.12. Follow-up analyses revealed that participants in the ‘phone present’ condition correctly solved significantly fewer matrices compared to participants in both the ‘phone absent’ and ‘experimenter phone present’ conditions (p<.05). The ‘phone absent’ and ‘experimenter phone present’ conditions were virtually identical (p>.05). These results are displayed in Figure 4-1. These data suggest that the presence of one’s own phone resulted in significantly poorer performance on a task of fluid intelligence compared to either the presence of another person’s phone or the absence of a phone. These data provide support for the notion that the presence of   68 one’s own phone impairs capacity to understand and engage with novel problems, corroborating Ward’s (2017) conclusions. This presents a successful first replication of Ward’s (2017) findings and by extension, provides support for the ‘brain drain’ hypothesis.  Notably, the presence of the experimenter's phone did not impair performance. Participants in the ‘experimenter's phone present’ condition were indistinguishable from participants in the ‘phone absent’ condition, which does not support Thornton’s (2015) findings that the presence of another person’s phone impaired performance on tasks that demand sustained attention, such as the RSPM.  The current study indicates that there is something unique about the presence of one’s own phone compared to the phone of another vis-a-vis its effect on fluid intelligence.  The aim of Experiment 2 was to try to pinpoint what that might be. In both the phone absent and experimenter phone conditions, there was no expectation that the phone could be acted on, e.g., to send or check messages. However, in the 'own phone present' condition, the expectation was clearly very different, as evidenced by the fact that some participants actually engaged with their phone during study. To determine if this factor is critical to the own-phone effect, we ran a second study, with participants being present with their phone when it was either turned on or off.   69  Figure 4-1 The effect of randomly assigned phone location on fluid intelligence, as measured by the RSPM. Participants in the “phone present” condition correctly solved fewer matrices than participants in the “phone absent” or “RA phone present” condition (p = .01). Error bars represent standard error of the mean.    4.3   Experiment 2 The purpose of Experiment 2 is to assess the phone-specific parameters under which phone presence is detrimental to performance on tasks of fluid intelligence.  Experiment 1 found that phone ownership matters when considering the effect of phone presence. That is, being in the presence of one's own phone is detrimental to performance, but being in the presence of another person's phone is not. Experiment 2 explores whether the ability for one's phone’s to signal and receive information (i.e. being turned on or off) is relevant to the effect of one's phone on cognitive performance.  As such, participants were randomly assigned to one of three groups (phone present, phone absent, or phone present but off) and asked to complete as many matrices from the Raven’s Standard Progressive Matrices (RSPM; Raven, Raven, and Court, 1998) as possible in 20 minutes.  16.9613.5217.2102468101214161820Phone  Absent Phone  Present RA  Phone  PresentCorrectly  Solved  Matrices  (out  of  25)  Phone  Location  70 4.3.1   Methods 4.3.1.1   Participants 69 undergraduates from the University of British Columbia (84% female; Mage= 20.0 years; SDage=3.55 years) participated in the experiment in exchange for course credit or monetary reimbursement.  Four participants were excluded: two misunderstand the task, one guessed the purpose of the study; and one had an acute mental health episode. Participants were randomly assigned to either the phone absent condition (n=22), the phone present condition (n=22), or the phone off condition (n=21). A reverse power analysis, based on Thornton (2014), revealed that the current study was powered at .60. 4.3.1.2   Materials See experiment 1.  4.3.1.3   Procedure The procedure for this experiment was the same as in Experiment 1, with the exception that the experimenter phone condition was replaced with the phone present (off) condition. The phone off condition was run identically to the phone present condition, with the exception that participants were instructed to turn off their phones for the entirety of the study.  4.3.2   Results and Discussion A univariate ANOVA revealed that there was no effect of phone condition on the number of correctly solved matrices, F(2,62)=1.94, p>.05. Participants performed similarly on the RPSM   71 regardless of whether their phone was absent or present, or when present, whether it was turned on or off 4. These results are displayed in Figure 4-2.   Figure 4-2 There was no effect of phone location on the number of correctly solved matrices. Scores across all conditions were comparable (p > .05). Error bars represent standard error of the mean.                                                  4 To eliminate the possibility that the failure to replicate the results from Experiment 1 was a result of low power (Experiment 2 was powered at .60), a mini follow-up study was conducted adding 10 participants to both the phone present (on) and phone absent condition. Even with this additional power (powered at .84), which was higher than Experiment 1 and higher than the Thornton (2015) study (powered at .67), the number of correctly solved matrices in the phone absent condition (M=16.81, SD=3.91) was not different than the number of correctly solved matrices in the phone present condition (M=17.58, SD=4.25), t(60) = -0.75, p = 0.46.    16.4117.5915.0502468101214161820Phone  Absent Phone  Present  (On) Phone  Present  (Off)Correctly  Solved  Matrices  (out  of  25)Phone  Location  72 In an effort to identify any other factors that could explain the discrepancy between results from Experiment 1 and Experiment 2, the only noteworthy finding is that on average, participants reported a 52 minute daily increase in time spent on their phone (Experiment 1 M = 171.3, SD = 92.4; Experiment 2 M = 222.8, SD = 178.1), a difference which trends towards significance, t(94) = -1.82, p = 0.07). Experiment 1 was conducted approximately one year prior to Experiment 2, suggesting that, on average, technology use had been increasing over time. The aim of Experiment 2 was to try to understand whether turning a phone off, and thus eliminating the expectation that the phone could be acted upon, could clarify why the performance decrement in Experiment 1 was only tied to self-owned phones. However, this decrement in performance relative to the phone absent condition was not observed, regardless of whether one's own phone was turned on or off. Indeed, if anything, having one's own phone improved performance when it was turned on. The aim of Experiment 3, therefore, was to understand why phone presence was shown to impair performance in some cases but almost boost performance in other cases. Specifically, the influence of a moderating variable that could explain both the replication failures and the contradictory effects was considered. 4.4   Experiment 3 Experiment 3 examines the potential moderating influence of prior technology use, or in other words, time spent online. ‘Time spent online’ has been identified as a key predictor for problematic internet use (PIU) (Cuhadar, 2012). In fact, there is a general consensus in the literature that the level of addiction in individuals who spend more than 5 hours (Odaci & Kalkan, 2010) or 8 hours ( Balta, & Horzum, 2008) online per day is higher than it is in individuals who spend less time online. If time spent online is so strongly associated with PIU, perhaps this variable will also be meaningful when considering the influence of mobile phone   73 presence on cognitive performance.  Given that mixed findings seem to be the norm within the small literature on cognitive phone presence effects, there is a theoretical rationale for at least two diverging predictions regarding the moderating effect of prior technology use. On the one hand, considering the prevailing ‘brain drain’ theory (Ward et al., 2017), it would stand to reason that individuals with higher levels of technology use would be more impaired by the presence of a phone compared to individuals with lower levels of technology use. This is because the phone would presumably be more salient to the individuals with higher levels of technology use, and therefore additional cognitive resources would be required to inhibit the influence of their phone. On the other hand, if cognitive performance were to mirror affective states, there is a case to be made for predicting that the performance of high technology users would actually be boosted by the nearby presence of their phone. For example, Cheever, Rosen, Carrier, and Chavez (2014) demonstrated that moderate and high technology users were more anxious than low technology users when separated from their phones. Also, Lyngs (2017) reported that high technology users had more fun during the experimental tasks when they were not separated from their phone. Intriguingly, although this is difficult to verify across studies given the absence of reported details on participant’s technology use, it is possible that Hartanto and Yang’s (2016) sample may have consisted of more avid technology users. They reported average smartphone use of 7.1 hours per day among their sample (which at the time was higher than average), and this was the same study that reported boosts in performance across the board.   In sum, the primary aim of Experiment 3 was to investigate and understand the mixed data generated not only within the present series of experiments but in the literature at large.   74 Specifically, the possibility of a moderating variable such as prior technology usage or ‘time spent online’ was explored. High and low technology users were recruited to participate in the study, and they were randomly assigned to either the phone present or phone absent condition.  4.4.1   Methods 4.4.1.1   Participants Participants were recruited based on their responses to a pre-screening questionnaire, where they were asked multiple choice questions to estimate their daily technology usage. See Appendix A for the exact questions. All undergraduates who were part of the Human Subject Pool were given the option of completing the pre-screening questionnaire which contained many questions submitted by various laboratories in the psychology department at the University of British Columbia. Importantly, participants did not know that they were recruited for this particular study as a result of their daily technology usage, thereby avoiding any demand characteristics (Boot, Blakely, Simons, 2011). Participants whose responses fell in the upper and lower quartiles of the distribution were invited to participate in the current study in exchange for course credit or monetary reimbursement. Participants were randomly assigned to either the phone present or the phone absent condition, and these groups were confirmed to be equivalent with respect to age and ethnicity. 162 undergraduates (70.4% female; Mage= 20.4 years; SDage= 3 years) participated. Their self-reported ethnicities were Asian (n = 76), Caucasian/White (n = 34), European (n = 5), Black (n = 3), Latin American (n = 2), Filipino (n = 4), Indian (n = 9), Middle Eastern (n = 5), Multiethnic (n = 7), and undisclosed/could not be categorized (n = 17). Five participants were excluded because they left the testing room prematurely and experimenters were unable to verify   75 the location of these participant’s phones. The final sample was thus composed of 157 participants. 4.4.1.2   A Note on Recruitment The decision to recruit participants from the upper and lower quartiles of the technology usage distribution was made with the intention of establishing a ‘high tech’ group and a ‘low tech’ group. However, the data obtained was unimodal and (M = 6.16 h/day, SD = 3.79, MED = 5.25) and normally distributed. If the recruitment methodology had been successful, I would have obtained a bimodal distribution. Araujo, Wonneberger, Neijens, and de Vreese (2017) and others (Scharkow, 2016) suggest that in addition to the usual challenges associated with accuracy of self-reported data, self-reports related to online behaviour and technology are even more difficult due to media multitasking and overlap across different platforms and devices. Perhaps unsurprisingly, then, it appears that despite being provided with identical questions during the pre-screen questionnaire and following the current experiment, participants were not consistent in their self-reports of technology use. Due to the failure of obtaining two distinct groups, I decided to move away from a 2x2 design and treat the technology use variable as continuous as opposed to binary.  4.4.1.3   Materials See Experiment 1.  4.4.1.4   Procedure The majority of participants were tested in groups in a classroom setting ranging between two and thirty participants. A small handful of participants (N=4) were tested alone. Each group of participants were randomly assigned to the phone present or phone absent condition. In the phone present condition, the group was instructed to place all of their belongings at the front of   76 the testing room, but to keep their phones with them during the study. A cover story that the phones would be needed at some point during the study was provided.  In the phone absent condition, the group was instructed to place all of their belongings at the front of the testing room, but to keep their student ID card with them. A cover story that the ID would be needed at some point during the study was provided. If a participant did not have their student ID card, they were told to keep another form of identification with them. Once the group was seated in the testing room, participants provided informed consent. They were then presented with verbal instructions for completing the RSPM. The experimenter worked through a sample item, and participants were given the opportunity to ask questions if they did not understand. Participants were then given 15 minutes to work through the package. They were told to complete as many as they could within this time frame, and the experimenter notified them when time was up. The experimenter and two research assistants walked around the room, monitoring the group as they completed this task as well as additional tasks unrelated to the current investigation. Participants were then presented with a detailed questionnaire assessing technology usage (see Appendix A for a copy of this questionnaire). They were instructed to complete the questionnaire independently, and were told that they would then be asked to review it with one of the experimenters. At this point, participants were directly asked about the location of their phone and this was verified by the experimenter. Participants were subsequently debriefed.  4.4.2   Results and Discussion An independent samples t-test revealed that there was no effect of phone condition on the number of correctly solved matrices, t(155)=0.955, p>.05. That is, as in Experiment 2, participants performed similarly on the RPSM regardless of whether their phone was absent or present. In contrast to the findings from Experiment 1, these data suggest that phone presence on   77 its own is not associated with decrements in performance on tasks of fluid intelligence. These results are displayed in Figure 4-3.                                                                                                                                                                  Figure 4-3 As was the case in Experiment 2, there was no effect of phone location on the number of correctly solved matrices. Scores across all conditions were comparable (p > .05). Error bars represent standard error of the mean.  Follow-up analyses were conducted to determine whether prior technology use would clarify the findings. A multiple linear regression was calculated to predict the number of correctly solved matrices based on phone location, technology use, and the interaction between phone location and technology use. A significant regression equation was found (F(3,153) = 3.40, p = .019), with an R2 of .063.  Participants’ predicted score on the RSPM is equal to 15.34 – 4.52 (Phone Location) – 0.34 (Technology Use) + 0.61 (Phone Location x Technology Use), where phone location is coded as 0 = Phone Absent, 1 = Phone Present, and technology use is 13.21 12.450510152025Phone  Absent Phone  PresentCorrectly  Solved  Matrices  (out  of  25)Phone  Location  78 measured in hours per day.  The interaction between phone location and technology use significantly predicted the overall RSPM score, b = .48, t(156) = 2.90, p = .004. These results are displayed in Figure 4-5.   Figure 4-5. A demonstration of how technology use and phone presence interact to predict the number of correctly solve matrices.   4.5   General Discussion This work was prompted by a desire to make sense of the mixed, and at times contradictory, findings regarding the influence of phone presence on cognitive performance. Some researchers have found evidence suggesting that the presence of a phone impairs performance on cognitive tasks (Thornton et al., 2014; Ward et al., 2017; Aguila, 2019), while others have found evidence that phone presence improves performance (Hartanto & Yang, 2016). 05101520250 2 4 6 8 10 12 14 16 18 20Number.of.Correctly.Solved.Matrices.(out.of.25)Technology.Use.(Hours/Day)Phone.AbsentPhone.Present  79 Others still propose that phone presence is a neutral factor (Lyngs, 2017).  We set out to understand why this literature contains so many failures to replicate, both across various cognitive domains (e.g., sustained attention, working memory capacity) as well as across and within different measures (e.g., additive digit cancellation task, trail making test).  After studying the methodologies employed by various researchers, it became apparent that there was considerable variability regarding how exactly the phones were presented to participants. In some cases, they were placed ‘face-up’ on the desk, and in others they were turned on silent mode. In others still, there was no mention of various phone parameters (i.e. placement, status, ownership, notifications) that could possibly be relevant to the question at hand. So, we set out to isolate some of these potentially key parameters in order to explore whether they might be significant when considering whether phone presence has an impact on cognitive performance.      Experiment 1 focused on the phone parameter of ‘ownership’ and found that phone presence only impaired cognitive performance when the phone was self-owned. When the present phone belonged to the experimenter, participant’s performance was indistinguishable from the condition where there was no phone present at all. This presents a successful replication of Ward (2017), who was the first to document that performance on tasks of fluid intelligence suffered as a result of phone presence. At the same time, these data contradict Thornton’s (2015) findings, which suggested that the presence of any phone, regardless of whether or not it was self-owned, hurt performance on tasks of sustained attention.  Experiment 2 focused on pinpointing what specifically made the difference between the other-owned and self-owned phone, and hypothesized that perhaps the other-owned phone was not associated with an expectation of being acted on or monitored for information. Thus,   80 Experiment 2 included a condition where participants' phones were present but turned off. Unexpectedly, the original effect from Experiment 1 was no longer detected. In light of the evidence that a phone presence effect may be somewhat fragile, I began to consider whether there might be a third variable influencing the data. I reasoned that perhaps the influence of a phone might be related to the relationship one already has with their phone. For instance, Hartanto (2016) obtained results suggesting that phone presence actually boosted performance, and he reasoned that being separated from one’s phone may actually be anxiety-provoking. We noticed that this study’s sample seemed to consist of more avid technology users, as they spent an average of 7.1 hours per day online, which at the time of the study was higher than average. We therefore turned our attention to the variable of ‘time spent online’, as we wondered whether phone presence might differentially affect individuals who were more avid technology users. Additionally, ‘time spent online' is also an excellent predictor for problematic internet use (Cuhadar, 2012), which we reasoned might also be a relevant factor.  Experiment 3 explored whether the amount of time an individual spends using technology (outside of the task itself) might help to clarify the relationship between performance on the Raven’s task and the presence of a phone. Interestingly, when the Experiment 3 data were analyzed in the same way as the data from Experiment 1 and 2 (that is, not addressing the influence of technology use), there was, once again, no difference in performance for participants who had their phones present compared to participants who had their phones absent. However, when the analysis took into account prior technology use, a clear pattern emerged, whereby phone presence assisted participants who were ‘high’ technology users and hurt participants who were ‘low’ technology users. That is, ‘high’ tech users performed better than ‘low’ tech users when they had their phones present. Similarly, ‘low’ tech users performed better than ‘high’ tech   81 users when they were separated from their phones. In other words, there was a significant interaction between the location of the phone and amount of prior technology use.  Ironically, the same issue that prompted this investigation (i.e. non-replicated data) presented itself within this very investigation. If we consider the trend in Experiment 2 showing that phone presence boosted performance, then we have three different findings herein: a) phone presence resulting in a decrement in performance in Experiment 1; b) phone presence trending towards possibly boosting performance in Experiment 2; and c) phone presence having no effect on performance in Experiment 3. The consideration of a third variable, which was chosen due to its ability to predict problematic internet use (Cuhadar, 2012), provided a key insight about the relationship between phone presence and cognitive performance. Based on these findings, high technology users should keep their phones with them when completing cognitive demanding tasks, whereas low technology users should keep their phones away. Certainly, these data raise several questions. For instance, why exactly is it that high tech users benefit from having their phones present while low tech users do not? Furthermore, does this effect extend beyond fluid intelligence to other cognitive domains? That is, would this unique impact of technology use on fluid intelligence task performance extend to other tasks such ones that utilize sustained attention, executive functioning, inhibitory control, or working memory? Perhaps the most interesting – and potentially consequential – question is this: aside from completing lab-based tasks specifically designed to measure fluid intelligence, does this effect actually exert influence outside of the lab? For instance, when high tech users are performing cognitively demanding real world tasks, such as doing their taxes or solving problems at work, do they actually make more mistakes when they are separated from their phones? If so, is this performance deficit salient enough that it significantly interferes with their   82 productivity or quality of life? Do they even notice? Are there functional impairments in the real world that are the result of either the presence or absence of mobile phones?  The present study cannot answer these questions, but at the very least, it sets a precedent for considering amount of technology use when investigating the influence of phone presence on behaviour. Furthermore, it provides an update to the popular “brain drain” hypothesis. Rather than merely being guided by the notion that a finite pool of cognitive resources is taxed by the presence of one’s phone, as Ward’s “brain drain” hypothesis posits, it may be more accurate to consider that our finite pool of cognitive resources might also be taxed by the absence of one’s phone, depending on whether one is a high or low technology user.  In conclusion, the impact of phone presence on cognitive performance is not straight-forward, and the literature is rife with mixed results. We posit that one of the reasons for these inconsistencies may be that the amount of prior technology use has not been considered as a key factor. We present the first evidence suggesting that rather than solely considering the variable of phone presence, considering the interaction between technology use and phone presence aids in understanding how performance on fluid intelligence tasks may vary. To the extent that the relation is causal rather than correlational,  present data suggest that high tech users should keep their phones with them when engaging in cognitively demanding tasks, while low tech users should put their phones away.        83 Chapter 5:  How do technology-related variables influence performance on cognitive tasks? An evidence-based response to popular fears 5.1   Introduction Whether and how the increasingly immersive web is impacting our cognitive abilities is currently a topic of intense and spirited public discussion. Indeed, mobile and related technologies are both ubiquitous – there are currently more online devices than people in the world (Cisco, 2014), and powerful – the average device boasts more computing resource than the Apollo 11 when it launched into space (Kaku, 2011). The present inquiry is largely motivated by mounting concerns about the impact that this era of connectivity is having on youths (Mills, 2016; George & Odgers, 2015) and on adults alike (Loh & Kanai, 2015). A quick glance at how previous generations have responded to new technologies reveals that panic and apprehension are not uncommon. Socrates, for example, advised his peers and students against writing things down as he believed that trust in this new invention would “discourage the use of their own memory” and therefore “produce forgetfulness” (Fowler, 1925). Similar fears emerged about how listening to radio programs, attending ‘moving pictures’, and reading comic books would damage children’s mental health (Heisler, 1948); and more recently, the invention of television (Maccoby, 1951) and later video games (Egli & Meyers, 1984) surfaced worries about how children’s development would be impacted. Concern over increasingly pervasive mobile and related technologies appears to be widespread and seems to stem from multiple sources including researchers, parents, educators, policy makers, and the popular press. For example, there are some researchers who claim that “screen based lifestyles” are “reducing the human desire to be inquisitive, think, comprehend,   84 and retain information” (Tripathi & Ahad, 2017) as well as “rewiring our brains” and creating a decline in higher order cognitive processes including inductive problem solving, critical thinking, and abstract vocabulary (Greenfield, 2009). Other researchers suggest that screen based lifestyles are impairing brain development such that the brains of millenials will be ‘wired’ for distraction instead of being ‘wired’ to stay on task (Richtel, 2010). Other still worry that the internet and mobile technologies may be altering reward and self-control mechanisms, and that the instant access to limitless information may be disrupting the utilization of effortful thinking. This is then hypothesized to reduce the need for deliberation and contemplation (Nasi & Koivusilta, 2012 as cited in Mills, 2009) and has implications for creativity as well. Several large-scale surveys have been conducted to investigate the beliefs that parents and educators have about the impact that hyperconnectivity has on youth. A survey of 802 parents of adolescents aged 12-17 revealed that 62% of parents have felt the need to regulate their child’s phone use by taking the phone away (Lenhart et al., 2015). Their concerns include fears about cyberbullying, worries about screen time detracting from engagement in “real life”, and apprehension about declining cognitive performance (George & Odgers, 2015; Madden et al, 2013; Boyd, 2014). In a book entitled, ‘Screen Schooled: Two Veteran Teachers Expose How Technology Overuse is Making Our Kids Dumber’, Clement and Miles (2017) claim that screen saturation has created a wide range of cognitive and social deficits in young people. These authors may not be in the minority: according to the Pew Research Center, 87% of high school teachers believe that widespread Internet use has created a ‘distracted generation with short attention spans’, and 88% felt that ‘today’s students have fundamentally different cognitive skills because of the digital technologies they have grown up with’. In addition, 64% of teachers polled reported   85 that today’s digital technologies “do more to distract students than to help them academically” (Lenhart et al., 2015).   Intriguingly, it is not just parents and educators who are concerned: in a survey about the future of the internet, 1,021 technology experts and stakeholders weighed in on their hopes and fears about the so-called hyperconnected generation. Forty-two percent of those polled believed that the overall impact of living hyperconnected lives would be “baleful”, and many of these respondents called to mind Orwell’s 1984, implying that younger generations are falling prey to the unhealthy ways of the internet just like the people of Oceania fall prey to the Party’s brainwashing  (Quitney & Rainie, 2012). Though 55% agreed with a statement suggesting that the future for those who are ‘always on’ will be mostly positive, many in this group predicted that there will also be negative repercussions including a decrease in both deep-thinking abilities and patience, as well as an increased propensity towards instant gratification.  With respect to policy, there have been concerted efforts from medical professionals, psychologists, and educators to develop national guidelines to restrict screen time for children in the UK (Palmer, 2016). The 40 signatories to this letter fear that without concerted action, children’s physical and mental health will “continue to deteriorate”, and that there would be social and economic repercussions for society in general.  Policy makers in the USA appear to be further along in that the American Academy of Pediatrics Council on Communications and Media has already developed broadly publicized policies around digital media use for very young children (AAP Council on Communications and Media, 2016a,b). The council asserts that early media use increases the likelihood of poor executive functioning in preschoolers. They therefore recommend limiting digital media use for children aged 2 to 5 years to no more than 1 hour per day and entirely avoiding digital media use for children younger than 24 months.   86 Unsurprisingly, the popular press also has much to say about the dangers of the internet (eg. Carr, 2010). Startling headlines, often appearing in the Science section, such as “Neuroscientist: The Internet is Destroying Your Brain” (Ehrenstein, 2012), “Is Google Making us Stupid?” (Carr, 2008), “8 Ways Tech Has Completely Rewired our Brains” (Hiscott, 2014), “Humans have shorter attentions span than goldfish, thanks to smartphones” (Watson, 2015). These and similar types of news reports paint the picture that the verdict is in and the evidence is compelling: technology is having a deleterious and lasting impact on our creativity and ability to think, reason, remember, and attend. What are we to make of these extensive concerns? Are they legitimate, or, as with many emerging literatures and new technological inventions (e.g., Socrates and “writing”), is opinion outpacing the data? Despite strong public sentiment that engagement with technology is damaging cognitive systems, the extant research supporting these claims is far from compelling. Though it is understandable that policy makers are eager to develop guidelines that prevent harm from befalling us, it would also be wise to make these decisions on the basis of empirically supported information. That is, in contrast to policies that are motivated by feelings and assumptions, those that emerge from a strong evidence base seek to reflect reality as accurately as possible. Data-informed positions have the benefit of offering a clear-eyed approach to decision making and if necessary, problem solving. On the other hand, solutions generated on the basis of inaccurate or biased beliefs may be ineffective and damaging. For example, the belief that smartphone usage impedes creativity resulted in a viral project called “Bored and Brilliant” where thousands of people vowed to abstain from checking their phones in the hopes that it would boost creativity (Zomorodi, 2015; as cited in Wilmer, 2017). To my knowledge, there is no empirical basis for the notion that smartphone usage stifles creativity, and though it is   87 possible that there is a negative link between smartphone use and creativity, it is also possible that the link is positive or simply nonexistent. In the latter cases, thousands of people would have unnecessarily altered their behaviour due to the influence of sensationalized information.  Despite it not presently being possible to make specific and confident claims regarding much of this literature, clear themes have emerged with respect to one unique area: the externalization of memory (i.e., relying on something external to oneself to remember). Externalizing memory is not a new concept – we use notepads to write important things down and we hang up our keys by the door so as not to forget them on our way out. Interestingly, there is now compelling evidence that the pervasiveness and accessibility of the online world is indeed influencing the strategies we utilize for remembering information. Specifically, there has been a shift towards using technological devices as an external source of memory. A ground-breaking study by Sparrow, Liu, and Wegner (2011) demonstrated that individuals tend to think of the internet essentially as an extension of their own brains and as a place to store and retrieve knowledge. For example, the authors showed that people are quick to think of computers in situations where they need new information. Furthermore, people are more likely to remember information that they think will not be available online compared to information that they think will be available online. Lastly, the authors demonstrated that people are also more likely to remember where a piece of information is stored online as opposed to remembering details about the item itself. These findings have been corroborated and supported through related inquiries into how memory may be changing in the digital age. For example, Ferguson, and McLean (2015) showed that individuals are less likely to volunteer answers to questions when they have access to the internet compared to when they do not have access to the internet. The authors suggest that this is because having access to a source of external memory means that information   88 is close by and can be checked easily – this makes it less likely for one to volunteer information without checking it. Fisher, Goddu, and Keill (2015) discovered an intriguing consequence to having the internet as a transactive memory partner: it appears to create the perception that one has more knowledge than is actually the case. Specifically, the authors found evidence that external information was mistaken as self-knowledge, suggesting that there can be a fuzzy line between what we consider to be stored in our own memory and what we consider to be stored online. Ward (2013) refers to this notion as the ‘merging of the minds’, where one’s own mind and the internet’s cloud mind blur together. Based on several clever experiments, Ward (2013) goes on to conclude that three main factors contribute to producing this effect: first, relating to the internet as a familiar transactive memory partner serves as an initial requirement; second, a “feeling of knowing” emerges; which leads lastly to an “I knew it all along” effect. As has been the case with externalizing memory, the current investigation seeks to contribute to the production of data-informed positions regarding other cognitive domains such as attention, impulsivity, executive functioning, and working memory. The current work joins many other scientific initiatives (e.g., Wilmer et al., 2017; George and Odgers, 2015; and Mills, 2016) in calling for a more measured and empirical based approach toward understanding the impact of an increasingly immersive web. A consolidation of the wide-spread concerns described earlier suggests that the main fears have to do with a belief that the current ‘hyperconnected’ era is producing people who are 1) distracted with shortened attention spans (Small & Vorgan, 2008; Lee et al, 2015; Carr, 2011), 2) impulsive and lacking in self-control (Giedd, 2012; Rich, 2010; Carr, 2011), 3) impaired in their ability to remember (Tripathi & Ahad, 2017; Greenfield, 2013), and 4) shallow, and impaired in their ability to think deeply and critically (Greenfield, 2009; Loh and Kanai, 2015; Giedd, 2012). To be clear, there are countless other important domains that are   89 certainly influenced by the ubiquity of mobile and related technology use, such as sleep, cyber-bullying, mental health, online safety, and the quality of social interactions and friendships (George & Odgers, 2015; Kushlev, 2015). These, however, are outside the scope of this investigation, as is an analysis of the studies addressing classroom-based learning and academic success in the context of laptop and mobile phone use (Sana, Weston and Cepeda, 2013). The current work considers both healthy populations as well as those that are thought to be pathological with respect to ‘excessive’ or ‘problematic’ use of technology. Unsurprisingly, research related to the idea that various forms of ‘internet use’ (e.g., video gaming, online gambling, social media use) may be associated with behavioural addiction-like symptomatology has burgeoned over the past fifteen years. Indeed, the 5th edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5; American Psychiatric Association, 2013) included ‘Internet Gaming Disorder’ in its section on emerging measures and models. Although internet-related addiction is not an official DSM-V disorder, many consider problematic internet use (PIU) (via online gaming or social media) to be related to deficits in impulse control (Beard & Wolf, 2001; Shaw & Black, 2008). Further, it is generally accepted that PIU can resemble addiction in that individuals may experience a decline in school or work performance, loss of interest in previously enjoyable hobbies, and conflictual relationships with the people around them (Kiraly, Nagygyorgy, Griffiths, & Demetrovics, 2014).  As an aside, concern about the impact of technology on cognition may in part be bolstered by shocking reports about individuals with PIU (e.g. cardiac-related deaths after video game players fail to eat and sleep: http://www.cnn.com/2015/01/19/world/taiwan-gamer-death/index.html). As Mills (2015) mentions, it is possible that this alarming press is skewing public perception of the negative effects of gaming and other online activities. PIU may be   90 relatively rare (Mills, 2016), and it seems unlikely that cognitive detriments experienced by those with the condition would be experienced to the same extent in people without the condition. Regardless, it is timely to explore how engagement with the online world may differentially affect individuals with PIU and those without, both for the sake of better understanding PIU itself and for understanding the impact of technology use more generally. If it is true that hyperconnectivity in general is influencing cognitive performance, it is reasonable to expect that those who spend inordinately more time online would see a greater disruption in cognitive responses. On the other hand, it is possible that like action video game players, people with internet addiction may experience cognitive benefits, such as enhanced attentional control (Chisholm & Kingstone, 2015). In addition to considering a spectrum of populations (i.e., ‘healthy’ to ‘pathological’), the current project focuses on the cognitive domains that are clearly implicated in both the extant literature and the public discourse, such as response inhibition, executive functioning, attention, impulsivity/reward processing, and working memory. Below I briefly summarize our current knowledge regarding the impact of technology use for each of these domains.  •   Response inhibition (the ability to suppress a behavior when needed). Previous research suggests that there does seem to be some emerging evidence for a negative relationship between technology use and inhibitory control, but this is mostly in the context of problematic gaming (Yao, 2015; Littlel et al., 2012) or other less-specified internet addictions (Nie, 2016).  Whether the impairment associated with performance is a result of the sheer magnitude of technology use or the ‘problematic’ or ‘addictive’ quality of the technology use remains unclear.    91 •   Impulsivity (a multidimensional construct, broadly defined as acting without adequate thought (Little, van der Berg, Luijten et al, 2012)). There is substantial evidence suggesting that pathological technology use is associated with a greater degree of impulsivity (Shin et al., 2019; Collins, Freeman & Chamarro-Premuzic, 2012; Mei et al, 2018). Relatedly, many studies converge on the finding that regardless of the type of problematic internet use (PIU) (e.g., gaming, social media), individuals in these populations tend to make riskier decisions compared to controls (Ko, 2010; Lin, 2015; Khoury et al, 2019).   •   Working memory (the capacity to hold information in mind and manipulate it (Strauss et al, 2014)). While there is some evidence for impaired working memory capacity in individuals with PIU (Nie, 2016; Zhou, 2016), data regarding media multitasking in healthy populations is very mixed. Some studies suggest that high media multitasking is associated with impaired working memory (Ralph & Smilek, 2016; Uncapher, Thieu, & Wagner, 2016), others suggest that that it confers a working memory benefit (Imren & Tekman, 2019), while others still claim there is no relationship between technology use and working memory (Gorman & Green, 2016; Seddon, Law, Adams & Simmons, 2018).  •   Attentional functioning. It should first be acknowledged that there is a large and established literature dedicated to describing the attentional benefits associated with playing action video games. Indeed, action video games players (AVGPs) have consistently been found to have better top-down attentional control, and as such, outperform non-AVGPs on tasks involving selective attention, sustained attention, and divided attention (Green & Bavelier, 2012). Regarding pathological technology use, there is some, albeit inconsistent, evidence for impaired attentional functioning (Dong, Zhou &   92 Zhao, 2011). As for healthy populations the data are mixed. Though many studies report that heavy media multi-tasking is associated with worse attentional performance (Beuckels et al, 2019; Ophir, 2009; Magen, 2017; Cain &  Mitroff, 2011), others suggest that heavy media multi-tasking confers an attentional benefit (Yap, 2013; Lui, 2012), while others still have reported that there is no relationship (Minnear, 2013, Wiradhany & Nieuwenstein, 2017).   5.2   The Present Investigation The main goal of the present investigation is to offer a more comprehensive understanding of how cognitive performance may vary as a function of technology use. Though there are a few trends beginning to emerge (e.g., the Internet has become a primary form of external memory (Sparrow, Liu, Wegner, 2011)), findings are, for the most part, still tentative and often mixed. Although this is to be expected given the relatively recent emergence of the immersive web as we know it today, there are a number of factors that make it difficult to answer the question at hand with the data that is currently available. First, there are many different methods employed (e.g. self-report, computer-based tests) which introduces variability when comparing across tasks. Second, there are many different populations (e.g. media multitaskers, problem gamers) under study, and rarely does the same study report data on more than one population, which introduces variability when comparing across populations. Third, studies that focus on healthy populations often seem to measure media-multitasking (which reflects the amount of media-multitasking one is engaged in, in a typical media consumption hour) as opposed to overall usage throughout the day. Fourth, many of the studies that focus on ‘internet addiction’ and problematic internet use often underspecify the type of technology use individuals are engaging in (e.g. social media, gambling, pornography etc.). It should also be noted that   93 problematic internet use (PIU) in general is a phenomenon that is itself not well understood, and for which there is no gold standard assessment tools. Instead, there are 21 different assessment tools, all with different cut-offs and based on different definitions of PIU (Kuss, Griffiths, Karila, Billieux, 2014). Lastly, although time spent online is a known predictor for PIU, the way that both of these variables (i.e., time spent online and PIU) interact to predict performance on cognitive tasks is unclear.   As such, the current investigation focuses on exploring how four variables relate to performance on cognitive tasks in the domains of attention, working memory, impulsivity, and executive functioning: 1) technology use (measured in hours per day), 2) level of problematic technology use (measured by the Internet Addiction Test), 3) phone presence, and 4) mental health symptomatology (measured by the SCL-90). This last variable was included given compelling evidence for a positive relationship between technology use and psychiatric disorders (Baer, Bogusz & Green, 2010) and general well-being (Twenge and Campbell, 2019). 5.3   Methods 5.3.1   Participants The participant sample was the same as in 4.4.1.1. 5.3.2   Procedure  The procedure was the same as in 4.4.1.2. Following completion of the Raven’s Standard Progressive Matrices (RSPM), participants were tasked with completing a battery of cognitive tasks and self-report measures, which were randomly presented on the computer.  5.3.2.1   Cognitive Tasks Continuous Performance Test- Identical Pairs (CPT-IP)   94  The CPT-IP measures sustained attention and inhibition by assessing the capacity to attend to continuously presented information (Rosvoid et al, 1956; Cornblatt et al., 1988). In this test, a series of stimuli flashes on the screen at a constant rate of 1 stimulus per second. The task requires participants to respond as fast as they can when two identical stimuli are presented in a row. My version used three stimulus conditions, which all participants were exposed to; numbers (N), numbers overlapped (NO), and numbers starred (NS). The dependent variable was a measure of sensitivity (d’), which was calculated by subtracting the rate of false alarms (responding when the stimuli were not identical) and misses (not responding when the stimuli were identical) from hits (responding, correctly, when the stimuli were identical). Trail Making Test (TMT) The TMT is a validated measure of attention and set-shifting (Reitan, 1958). The test consists of two parts and requires alertness and concentrated attention to complete. In Part A, there are 25 circles which contain numbers (1-25), and participants must draw one continuous line to connect the numbers in ascending order. In Part B, there are again 25 circles which now consist of both numbers (1-13) and letters (A-L), and participants are required to draw a continuous line alternating between ascending numbers and letters. The dependent variable was the elapsed time required for the task, calculated in milliseconds. Automated Operation Span (OSPAN) task The OSPAN task is a measure of cognitive control and executive functioning that assesses the capacity of task-relevant information during complex cognitive tasks (Turner and Engle, 1989). In this task, participants are presented with a sequence of 3-7 letters to be recalled at the end. Each letter is followed by a math problem with a proposed solution, and participants   95 are instructed to indicate whether the solution is correct or incorrect. Accuracy of recall is tested later by asking participants to select the sequence of letters from a matrix.  Balloon Analog Risk Task (BART) The BART is a measure of impulsivity and risk taking behaviour (Lejuez et al., 2002). In this task, individuals view an animation of a balloon, which they are instructed to click on to inflate. Each click earns ‘money’. The balloon will pop at some unknown threshold, thus making each pump click a risk. An individual earns more money each time they inflate, but if the balloon pops before they cash out, they will lose the money from that trial. The dependent variable is the average pump count.  Digit Span The Digit Span is a validated measure of working memory (Lumiley & Calhoon, 1934). I utilized an auditory backwards version of the Digit Span, where participants were presented with an auditory list of words and then instructed to recall the words in reverse order.  Single N-back task  The Single N-back is a measure of spatial working memory (Owen et al., 2005). Participants were shown a sequence of shapes and told to indicate whether the currently presented stimuli was identical to the shape of an earlier presented stimulus. N=0 trials required participants compare the current stimuli to the one immediately preceding.  N=1-4 trials require participants to indicate if the current stimuli is equivalent to the one presented right before the preceeding one (N=1), two before the preceeding one (N=2), three before (N=3), or four before (N=4).  5.3.2.2   Self-report Measures Symptom Checklist 90 (SCL-90)   96 Self-reported mental health was assessed using the SCL-90, which measures the presence and intensity of psychological distress over 90 items (Sereda & Dembitskyi, 2016). The primary symptom dimensions include somatization, obsessive-compulsive symptoms, interpersonal sensitivity, depression, anxiety, hostility, phobic anxiety, paranoid ideation, psychoticism, and additional items which contribute to the global severity index (GSI), which is the global index of distress. Internet Addiction Test (IAT) The 20-item IAT was used to measure the presence and severity of Internet dependency among participants (Young, 2001). It is the most frequently used measure of internet addiction and has been shown to have good validity, reliability (Widyanto & McMurran, 2004), and satisfactory internal consistency (Kuss, Lopez-Fernandez, 2016). Technology Usage  Participants were then presented with a detailed questionnaire assessing technology usage (see Appendix A for a copy of the Technology Usage questionnaire). Data for the variable ‘Social Technology Use’ was derived by summing the values for texting, messaging, social media, and email. The variable ‘Gaming Technology Use’ was derived from the values for gaming. ‘Total Technology Use’ was derived by summing all of the types of technology use shown in Appendix A. Participants were instructed to complete the questionnaire independently, and were told that they would then be asked to review it with one of the experimenters. At this point, participants were directly asked about the location of their phone and this was verified by the experimenter. Participants were subsequently debriefed.     97 5.4   Results and Discussion 5.4.1   A Note Regarding Chosen Analyses I consider the following analyses to be exploratory in nature. Although a number of these comparisons have been made before (e.g., the influence of phone presence on the trail-making test), the approach has generally been piecemeal, such that it remains difficult to understand the broader context in which performance on cognitive tasks is influenced by technology-related variables. It is for this reason that I employ the use of several multiple linear regressions. Because this approach runs the risk of increasing a Type 1 error (i.e., rejection of a true null), the conclusions I make are tentative. However, I believe this approach is appropriate because the purpose of the current investigation is to survey the landscape and understand which technology-related variables may be relevant when considering a number of different cognitive domains.  Regarding the assumption of independence for each of these regressions, as I indicate below, there are indeed moderate correlations between some of the predictors (e.g., IAT score and technology scores). However, because the highest correlation is .41, and none of the Variance Inflation Factors (VIF) exceed 10 (the highest VIF is 4.94), I can conclude that there are no issues with multicollinearity (Hair et al., 2014) and can proceed with the regressions.        98 5.4.2   Correlations for Main Study Variables       Total  Technology Use Social Technology Use Gaming Technology Use IAT Score IAT Score .41** .34** .12 ns  GSI Score .26** .19* .11 ns .39**  Table 5-1 Pearson correlations for main study variables.   IAT = Internet Addiction Test, GSI = Global Severity Score, ns = not statistically significant, * = statistically significant at the p <.05 level; ** = statistically significant at the p < .01 level.  Table 5-1 outlines the correlations of interest for the main study variables. There was a statistically significant, moderate positive correlation between daily total technology use and IAT score, with daily total technology use explaining 16% of the variation in IAT scores. Interestingly, this relationship was driven by social technology use (i.e., texting, messaging, social media use, and email) and not by gaming technology use, as the latter was not correlated with IAT score, while social technology use explained 12% of the variation in IAT scores. This suggests that, as social technology use increases, so does one’s score on the IAT. Figure 5-1    displays these relationships. Table 5-2 outlines the descriptive statistics of interest for the main study variables.     99    CPT d'  TMT (ms) Digit Span  OSPAN  BART IAT GSI Mean   0.6166   7.005e+4   6.046  47.25   29.71   35.29   63.85   Std. Deviation   0.2321   2.293e+4   1.344  16.67   13.96   16.17   10.48   Minimum   0.000   3.157e+4   1.000  7.000   3.133   0.000   32.00   Maximum   0.9670   1.271e+5   10.00  75.00   62.13   79.00   80.00    Table 5-2 Descriptive statistics of interest for the main study variables.       100                 Figure 5-1. The relationship between technology use (total, social, and gaming) and score on the Internet Addiction Test5  To further analyze the IAT scores of gamers, a one-way ANOVA was conducted. Specifically, to investigate whether IAT scores differed as a function of gaming frequency, the gaming technology use variable was binned according to how varying levels of gaming are defined in the literature (Chisholm & Kingstone, 2015; Luijten et al, 2015). As such, the cut-offs were 28 or more hours per week (heavy gamers), 8-27 hours per week (gamers), or fewer than 8 hours per week (non-gamers). There was no main effect for gaming frequency, F(2, 155) = 1.36, p = .26, suggesting that IAT scores are similar for heavy gamers, regular gamers, and non-gamers. One possible explanation for this result is that social technology use may be offsetting, or masking, the effect of gaming (e.g., gamers may be using social technology more than heavy gamers).                                                 5  The  correlation  between  Gaming  Technology  Use  and  IAT  score  remains  non-­‐significant  if  the  non-­‐gamers  (i.e.  0  hours  of  gaming  per  day)  are  removed  from  the  analysis.    101 There was also a statistically significant, small positive correlation between daily total technology use and GSI score, with daily total technology use explaining 7% of the variation in GSI score. Interestingly, this relationship was once again driven by social technology use and not by gaming technology use, as the latter was not correlated with GSI score, while social technology use explained 4% of the variation in GSI score. Figure 5-2 graphically displays these relationships.     102   Figure 5-2. The relationship between technology use (total, social, and gaming) and score on the Global Severity Index.    103 5.4.3   Attention & Inhibition A multiple linear regression was run to predict performance on the Trail Making Test (TMT) from technology use, IAT score, phone location, GSI score, and the interaction between phone location and technology use. The model did not predict TMT performance, F(5, 139) = .178, p = .79, adj. R2 = -.029. Performance on the TMT was not influenced by degree of technology use, degree of internet addiction, phone location, or mental health symptomatology. This suggests that individuals’ ability to shift attention was not impaired or aided by any of these variables. This is noteworthy, as Thornton (2014) was the first to report a cognitive impact for phone presence: in fact, this study concluded that TMT performance was impaired by the presence of one’s smartphone. The sample size for this study (n=37) was much smaller than that of the current study (n=157), so it is possible that Thornton (2014) detected a fragile effect or that there are key methodological differences between the two studies. It should be noted, however, that Aguila (2019) (n=56), also failed to find an effect of phone presence on TMT performance. A multiple linear regression was run to predict performance on the CPT-IP from technology use, IAT score, phone location, GSI score, and the interaction between phone location and technology use. The model did not predict CPT-IP performance, F(5, 139) = 1.03, p > .58, adj. R2 = .001. Performance on the CPT-IP, in terms of d prime (sensitivity) was not influenced by degree of technology use, degree of internet addiction, phone location, or mental health symptomatology. There was a statistically significant, small positive correlation between the number of random errors (i.e. responses made during filler trials, when the stimuli presented had no digits in common with the preceding stimuli) and score on the IAT, r(142) = .180, p =   104 .03, suggesting that individuals with higher levels of internet addiction made more random errors.  5.4.4   Executive Functioning / Cognitive Control A multiple linear regression was run to predict performance on the OSPAN from technology use, IAT score, phone location, GSI score, and the interaction between phone location and technology use. The model did not predict OSPAN performance, F(5, 148) = .222, p > .93, adj. R2 = -.027. Performance on the OSPAN was not influenced by degree of technology use, degree of internet addiction, phone location, or mental health symptomatology. This suggests that individuals’ cognitive control was not impaired or aided by any of these variables.  Figure 5-3 illustrates the relationship between technology use and OSPAN performance as a function of phone location (i.e. absent or present). As shown in the figure, OSPAN performance remains remarkably consistent regardless of the amount of technology use and the location of the phone. This finding contrasts with Ward (2017), who was the first to report that phone presence impaired performance on the OSPAN task. Although the current samples size (n=157) is large for phone presence studies (sample sizes range from n=37 to n=87), Ward (2017) employed the use of a relatively very large sample size (n=297), which would have been powered to detect even very small effects. Figure 5-4 illustrates the relationship between IAT score and OSPAN performance as a function of phone location (i.e. absent or present). As shown in the figure, once again, OSPAN performance remains remarkably consistent regardless of IAT score and the location of the phone. This finding also contrasts’ with Ward (2017), who reported that higher degrees of phone dependence resulted in more performance impairment when the phone was present.    105                Figure 5-3. The relationship between technology use (total, social, and gaming) and OSPAN score as a function of phone presence.   PresentAbsent  106     Figure 5-4. The relationship between internet addiction and OSPAN score as a function of phone presence.   Despite there being an absence of technology-related effects on the OSPAN task, there did seem to be significant effect of phone presence, t(148) = -2.38, p = .018, pertaining to the number of errors made on the filler task (i.e., the mathematics problem that was presented after the initial string of letters and before the recall portion of the task), such that individuals who had their phones present made more errors than individuals who did not have their phones (see Figure 5-5).  PresentAbsent  107          Figure 5-5. The relationship between internet addiction and OSPAN score as a function of phone presence.   5.4.5   Impulsivity A multiple linear regression was run to predict performance on the BART from technology use, IAT score, phone location, GSI score, and the interaction between phone location and technology use. The multiple regression model predicted BART performance, F(5, 148) = 2.50, p = .03, adj. R2 = .048. As shown in Figures 5-6 and 5-7, the analysis revealed that degree of technology use, (B = -.30, t(142) = -2.70, p = .01), and GSI score, (B = -.17, t(142) = -1.95, p = .05) significantly predicted the adjusted BART pump count. Degree of internet addiction and phone location were not predictive of the pump count. These findings suggest that as technology use increases, impulsivity and risk tolerance (i.e., willingness to pump the balloon) decrease (Figure 5-6). Regarding ratings on the GSI, further analyses demonstrated that depressive symptoms were 3.515.070123456Phone/Absent Phone/PresentOSPAN&Errors  108 driving the effect of GSI score on pump count. So, it seems that as depressive symptomatology becomes more severe, risk tolerance decreases (Figure 5-7).            Figure 5-6. The relationship between technology use and adjusted pump count on the BART.     109  Figure 5-7. The relationship between depression symptomatology and adjusted pump count on the BART.  When technology use was broken down into social technology use and gaming technology use, the small, statistically significant negative correlation between pump count and social technology use persisted, r(148) = -.21, p < .01, suggesting that increasing use of social technology was associated with lower risk tolerance (Figure 5-8). In contrast, there was a small, statistically significant positive correlation between pump count and gaming technology use, r(148) = .17, p = .04, suggesting that as gaming increases, so does risk tolerance (Figure 5-9).    110           Figure 5-8. The relationship between social technology use and adjusted pump count on the BART.   Figure 5-9. The relationship between gaming technology use and adjusted pump count on the BART.   111  5.4.6   Working Memory A multiple linear regression was run to predict performance on the Digit Span task from technology use, IAT score, phone location, GSI score, and the interaction between phone location and technology use. The multiple regression model did not predict Digit Span performance, F(5, 150) = 1.67, p > .05, adj. R2 = .022. However, as shown in Figures 5-10, the analysis showed that degree of technology use, (B = -.22, t(144) = -2.02, p = .046) significantly predicted Digit Span performance. That is, as technology use increased, performance on the digit span worsened. This same pattern was observed for social technology use, r(148) = -.14, p = .08, but there was no correlation between gaming technology use and digit span performance, r(148) = .001, p > .05.  Figure 5-10. The relationship between technology use and digit span performance.   112  A multiple linear regression was run to predict performance on the N-back task from technology use, IAT score, phone location, GSI score, and the interaction between phone location and technology use. The multiple regression model predicted N-back performance, F(5, 149) = 3.30, p < .01, adj. R2 = .07. As shown in Figure 5-11, the analysis showed that IAT score, (B = -.29, t(143) = -3.12, p < .01) significantly predicted the N-back score. This suggests that as the degree of internet addiction increases, performance on the N-back score worsens. Phone location, technology use, and mental health symptomatology were not significant predictors of N-back performance.   Figure 5-11. The relationship between IAT score and N-back performance.   113 5.5   General Discussion The current investigation was prompted by mounting concerns about the impact that our era of connectivity is having on youths (Mills, 2016; George & Odgers, 2015) and on adults alike (Loh & Kanai, 2015). These concerns are widespread and broadly shared by researchers, parents, educators, policy-makers, and even stakeholders in the tech industry. The fears have to do with a belief that the current ‘hyperconnected’ era is producing people who are 1) distracted with shortened attention spans (Small & Vorgan, 2008; Lee et al, 2015; Carr, 2011), 2) impulsive and lacking in self-control (Giedd, 2012; Rich, 2010; Carr, 2011), 3) impaired in their ability to remember (Tripathi & Ahad, 2017; Greenfield, 2013), and 4) shallow, and impaired in their ability to think deeply and critically (Greenfield, 2009; Loh and Kanai, 2015; Giedd, 2012).  Unfortunately, opinion is outpacing the data. The relevant scientific literature, although burgeoning, is not established enough to offer strong and empirically-based direction regarding the ways in which an increasingly immersive web is impacting cognitive systems. Despite the intense public sentiment that engagement with technology is damaging cognitive systems, the extant research supporting these claims is far from compelling. Accordingly, the goal of the present investigation is to organize and summarize the current understanding of the ways that technology-related variables influence performance on cognitive tasks, and then to extend that knowledge.  A summary of the findings from the current investigation is shown in Table 5-3. Two of the more surprising findings were that phone location never influenced task performance (in contrast to Ward et al., 2017; and Thornton et al., 2014), and that IAT score was largely unrelated to task performance for the impulsivity domain (Lin, 2015). That said, it should be noted that the majority of impulsivity studies in pathological populations utilized self-report   114 measures as opposed to behavioural tasks (Shin et al, 2019; Kim, Namkoon, Ku, & Kim, 2008; Collins, Freeman, & Chamarro-Premurzic, 2012). In fact, there was absolutely no evidence herein that attentional performance, response-inhibition abilities, or executive functioning were related to any of the technology-related variables.    Performance Predictors  Cognitive Domain  Technology Use  IAT Score  Phone Location  GSI Score Task Total Social Gaming  Attention & Inhibition TMT No No No No No No CPT-IP No No No No* No No Executive functioning OSPAN No No No No No+ No  Impulsivity & Reward Processing   BART Yes: Higher tech use, lower risk tolerance Yes: Higher tech use, lower risk tolerance Yes: Higher tech use, higher risk tolerance No No Yes: Higher depressive symptomatology, lower risk tolerance    Working Memory  Digit Span Yes: Higher tech use, worse working memory Yes: Higher tech use, worse working memory No No No No  N-Back No No No Yes: Higher IAT score, worse working memory No No Table 5-3. Summary of findings.  *High IAT scores associated with more random errors on CPT-IP +More filler errors were made when phones were present   The typical approach among these studies has been more piece-meal than wholistic: experiments typically focus on one or two cognitive domains and one population (i.e. media-multitaskers), and often underspecify the type of technology use. This makes it challenging to   115 disentangle the variables that contribute to either impaired or enhanced performance. Though the overlap between time spent online and negative consequences related to technology use remains unclear, to my knowledge, the present study provides the first demonstration of how social versus video-game related types of technology use differentially predict task-performance in the domains of working memory and risk. For instance, on digit span, higher levels of social technology use were associated with fewer recalled words, while levels of gaming technology use were not predictive of task-performance. Additionally, on the BART, higher levels of social technology use were associated with a lower tolerance for risk, while the opposite was true for gaming technology use, where higher levels were associated with riskier decisions. This result is intriguing because it suggests that different types of technology use may have different effects on the likelihood of making risky or impulsive decisions. Similarly, it could be that gaming technology use is not associated with decrements in working memory, but that social technology use is. Lastly, these results suggest that internet addiction symptoms may influence spatial working memory (N-back) but not auditory working memory (Digit Span).  A big limitation of this work has to do with the inability to understand the temporality of the effects that did emerge. For instance, did high levels of technology use result in higher impulsivity levels or do impulsive individuals find themselves more likely to engage in more frequent online behaviours? There is some evidence to suggest the latter – Gamez-Guadix, Calvete, Orue, & Las Hayas (2015) conducted a longitudinal study on a population of adolescents aged 13 to 18 years. The adolescents’ ability to self-regulate online usage predicted the degree of negative technology-related consequences they later experienced. A related limitation, as mentioned previously, is the difficulty with disentangling amount of time spent   116 online with addiction-related symptomatology. For example, it was unclear whether self-regulation abilities were also associated with more time spent online in the absence of negative consequence. When Cain and Mitroff (2011) found that heavy media multitaskers had greater attentional deficits compared to light media multitaskers, they put forth an interesting possible explanation for their findings. They suggested that people who are naturally more distractible may be more likely to surround themselves with multiple forms of media so that when they do inevitably get distracted, they are intentionally distracted by media they enjoy as opposed to being interrupted by unenjoyable distractions. Regardless, given issues of causality, it is not possible to know for sure whether it is actually the high levels of technology use in this study that resulted in lower working memory for high social tech users, for example. There is of course the possibility that those with pre-existing cognitive deficits go on to become more frequent technology users or media multitaskers. It is easy to imagine, also, that there could be a dynamic interplay between these variables, such that pre-existing cognitive deficits may predispose a person to becoming ‘addicted’ to online behaviours which then reinforces those cognitive deficits in some way.  As previously noted, given the exploratory nature of the analyses, the above findings and conclusions should be considered tentative and in need of replication. Though it is still not possible to provide specific and confident responses regarding the cognitive concerns associated with the increasingly immersive web, it is encouraging to note that especially among healthy populations, the evidence as a whole does not suggest substantial technology-induced cognitive impairment. However, more research is necessary before firm conclusions about the effects of technology on cognitive impairment can be drawn.    117 Chapter 6:  General Discussion My doctoral work was born out of a desire to cultivate and contribute empirically based insights about important issues of our time. This document is centered on two main issues that present both challenges and opportunities for our global cultural moment: environmental sustainability and the increasingly immersive web. In the current chapter, I provide a brief summary of the key findings and implications pertaining to each of the chapters, followed by a discussion of the data in the context of the dissertation’s overarching frameworks: situated cognition and nudge theory. Subsequently, I consider the overall implication, limitations, and conclusions related to the present body of work.  6.1   Chapter Summaries Chapter 2 illustrates how a simple intervention reveals important implications for waste management and environmental policy. In two randomized field experiments, I demonstrate that convenience dramatically boosts recycling and composting rates in multi-family dwellings and university residences. When compost bins were placed on each floor in a multi-family residence instead of on the ground floor, composting rates increased by 70%, diverting 27 kilograms of compost from the landfill per unit per year. When recycling stations were placed just meters from suites in student residences instead of in the basement, recycling increased by 147% (container), and 137% (paper), and composting increased by 139%, diverting 23, 22, and 14 kilograms of containers, paper, and compost, respectively, from the landfill per person per year. Simply by making recycling and composting convenient, it is possible to nudge individuals towards increasing the frequency of sustainable behaviours, and consequently increase waste diversion.   118 Chapter 3 reveals that subtle variations in context can modify the extent to which pro-social behaviours are displayed. It is well known that the presence of other people modifies behavior. In fact, even an implied social presence, as opposed to an actual social presence, has been shown to influence looking behavior. For instance, wearing an eye tracker has been shown to reduce a wearer’s willingness to look at provocative images. In Chapter 3, I presented the first evidence that the presence of an eye tracker can also encourage pro-social behavior beyond looking behaviour itself. Not only were the participants who wore an eye tracker less likely to cheat on a puzzle task compared to those without an eye tracker, those with an eye tracker were also more likely to confess whenever they did cheat. This finding has implications both for the way that laboratory research is conducted and for consumers and manufacturers of wearable computing. Indeed, the research in this chapter demonstrates that countless experiments may have unwittingly been based on participants behaving more pro-socially than they usually do, and suggests that there may be pro-social benefits associated with smart glasses or augmented reality (AR) goggles. Chapters 4 demonstrates that the presence of mobile phones is not a neutral factor when it comes to performance on tasks of fluid intelligence. A very small body of research has previously explored the cognitive impact of the presence of smartphones, but the outcomes of that work are inconclusive and contradictory. Some researchers have found evidence suggesting that the presence of one’s phone impairs performance on cognitive tasks (Thornton et al., 2014; Ward et al., 2017; Aguila, 2019), while others have found evidence that phone presence improves performance on cognitive tasks (Hartanto & Yang, 2016). Others still propose that phone presence is a neutral factor (Lyngs, 2017).  In Chapter 4, I sought to clarify the influence that the presence of a smartphone has on tasks of fluid intelligence by a) experimenting with   119 phone-specific parameters and b) investigating whether the degree of prior technology use may be a moderating factor. For high technology users, scores on a test of fluid intelligence were higher when the participants’ smartphones were present. For low technology users, scores were higher when the participants’ smartphones were absent. Collectively these data suggest that when working on a task requiring fluid intelligence, high users should keep their phones with them, and low users should put their phones away. Moreover, without taking into account the impact of prior technology use as a moderating variable, interpretations about the impact of phone presence are limited.  These findings have implications for how we think about the effects of an increasingly immersive web, both with regard to personal factors (e.g., time spent online) and situational factors (e.g., presence of one’s phone). Chapter 5 extends the work of Chapter 4 such that performance in a number of different cognitive domains were studied in the context of technology use, mobile phone presence, mental health symptomatology, and internet addiction. This chapter highlights the reality that although some trends are emerging, there remain many unknowns regarding the influence of ubiquitous connectivity on human cognition. For instance, unlike in Chapter 4, where phone presence significantly predicted performance on tasks of fluid intelligence, in Chapter 5 phone presence did not predict performance on tasks related to attention, inhibition, executive functioning, working memory, or impulsivity. In fact, none of the proposed predictors had reliable influences across cognitive domains. However, technology use was shown to be related to lower impulsivity and worse working memory performance. Additionally, social versus gaming technology use consistently predicted different outcomes (e.g., the former was associated with lower tolerance to risk and the latter was associated with riskier decisions).   120 6.2   The Research With Regard to Situated Cognition and Nudge Theory I consider the frameworks of situated cognition (i.e., cognitive processes are intimately coupled with the surrounding environment as opposed to being invariant and stable, e.g., Kingstone, Smilek, & Eastwood, 2008) and nudge theory (i.e., the notion that behaviour can be dramatically influenced by subtle changes in context) as lenses through which to orient, understand, and consider the broader work. Further, I submit that using these frameworks is useful as it allows for the uncovering of effects that may have otherwise been missed. As mentioned in Chapter 1, despite substantial overlap between the two theoretical frameworks, this work presents the first scenario in which situated cognition and nudge theory are considered in relation to each other. They are both firmly grounded in the assertion that context is significant and should be factored into decisions and conclusions related to the fields of cognitive science and behavioural economics. By way of comparison, I have suggested that while situated cognition is primarily concerned with description, nudge theory is usually associated with a particular application related to a desired outcome. That is, situated cognition is concerned with accurately understanding a cognitive process within a particular context, and nudge theory is concerned with solving a problem within a particular context.  With respect to the ways that these frameworks could provide additional explanatory power, let us first consider the findings from data Chapter 2. For instance, the sustainability literature has generally overlooked the ways in which environmental changes (i.e., changes in physical context) can alter the likelihood of engaging in sustainable behaviour. Despite emerging evidence that changes to the context can improve sustainable behaviour on an aggregate level (Wu, DiGiacomo, Kingstone, 2013), the field is still mostly focused on individual traits (e.g., improving education) and values (e.g., appeals to morality) that could positively influence pro-  121 environmental behaviours (Osbaldiston & Schott, 2012). This approach assumes that human beings have ready and reliable access to the Reflective System (Kahneman, 2012), and underestimates the benefits of circumventing this system by creating an environment that promotes sustainable behaviours even when the Automatic System is being used. Indeed, the data from Chapter 2 suggest that I was successfully able to nudge residents to recycle and compost more frequently by moving recycling and compost bins closer to their suites. Although the framework of situated cognition is absolutely relevant in this case (e.g., recycling rates depend on the proximity of recycling bins), the additional use of the nudge framework is most appropriate because there was a specific intention (e.g., to increase recycling rates) and a clear position (e.g., it is better to divert waste from the landfills by recycling/composting than to add waste to the landfills) guiding the intervention.  With respect to the remaining data chapters (Chapters 3 through 5), I submit that for the purposes of this dissertation, the situated cognition framework is more appropriate than the nudge theory framework. However, I acknowledge that opinions regarding this may vary and suspect that this may depend on the weight and importance one ascribes to the presence of ‘unintended nudges’ within the nudge theory framework. Though it is clear that Thaler and Sunstein (2008) leave room for unintended nudges by suggesting that people may not be aware of their position as ‘choice architects’, I take the position that these instances of ‘unintended nudging’ are more peripheral and do not reflect the heart of nudge theory -- which is to influence behaviour in a particular direction. I would argue that ‘unintended nudges’ are almost identical to situated cognition, with the exception that they may not occur in the context of attempting to accurately describe a cognitive process. Considering data from Chapter 3, proponents of wearable computing have not considered the possibility that merely wearing smart glasses or   122 augmented reality (AR) goggles might yield benefits or repercussions (depending on the perspective). Researchers who employ eye tracking technologies have been unaware that participants in eye tracking experiments may behave in a profoundly different way simply because they are wearing an eye tracker. Considering data from Chapters 4 and 5, those concerned with the effects of an increasingly immersive web may not have considered the possibility that in addition to the over-use of devices, being in the presence of mobile technologies may have consequences for cognitive systems. Rather than taking a moral position on whether wearable computing is beneficial or whether working in the presence of a phone is preferable to working in the absence of a phone, these data merely put forward the reality that frequency of cheating or performance on (some) cognitive tasks is affected by the specific context participants find themselves in. That is, the cognitive processes underlying the tasks were dynamic and varied according to the context. This is precisely the type of discovery that situated cognition is concerned with.  6.3   Implications The current dissertation showcases the utility of operating within a situated cognition framework. To borrow an illustration from Wilson and Clark (2009), consider the case of the hermit crab. When a hermit crab finds an empty shell on the beach, it uses the shell for protection and shelter. The hermit crab and the shell together, therefore, possess resources and abilities that the hermit crab would not possess on its own. The shell can be seen as an extension of the hermit crab’s cognitive system, such that it modifies the crab’s cognitive behaviours and activity. In order to glean a comprehensive understanding of cognitive processes, then, it is not sufficient to merely study what goes on within the mind, so to speak. That is, an examination of the hermit crab without its shell would only tell part of the story. A fuller, and perhaps richer picture   123 emerges when reciprocal transactions between the mind, body, and world are considered (Robbins & Aydede, 2009).  The situated cognition framework can therefore be regarded as a dynamic and fluid investigation of cognitive extensions into both the physical and social world (Wilson & Clark, 2009). These ‘extensions’ will vary widely, both in their durability and their form, and in the degree to which we are aware that they are being incorporated into our cognitive work. For instance, the shells on the beach are consistently and reliably available to the hermit crab, who has ‘learned’ to rely on the shell (i.e., offload its internal cognitive load) for protection. In a similar way, the conveniently placed bins in Chapter 2 serve as a cognitive extension. Their presence allows for individuals to recycle and compost much more frequently, and therefore behave in accordance with their own values (as most of those individuals reported that they believe composting and recycling are important). Like the shell is a natural extension of the crab’s cognitive system, the constancy of the conveniently placed bins become an extension of the building residents’ cognitive system. Both the shell and the bins serve to lighten the cognitive load of the hermit crab and building residents, respectively. Clark (1989, p.64) explained this notion by saying that “evolved creatures will neither store nor process information in costly ways when they can use the structure of the environment and their operations upon it as a convenient stand-in for the information processing operations concerned”.  Not all cognitive extensions, however, are as permanent or durable as the shell or the bins. For example, consider the eye tracker wearers in Chapter 3. The combination of the participants and the eye tracker together resulted in more pro-social behaviour and a higher degree of honesty. In this case, unlike in Chapter 2, it does not make sense to conceptualize the situated cognition environment as one that allows for cognitive off-loading. Rather, the   124 introduction of the eye tracker into the participant’s environment could have initialized a mechanism that underlies the attention theory hypothesis (Carver & Scheier, 1981). For instance, as Guerin (2010) posits, individuals who feel observed (regardless of whether or not they are actually being observed) may respond by shifting their attention toward themselves. This corresponding increase in self-awareness is thought to bring about social presence effects, often consistent with impression management. Regardless, the methodological implications resulting from a situated approach to this research are worth noting. After all, the new knowledge that eye tracker presence could influence pro-social behaviour directly emerged from the decision to contrast behaviour from a ‘typical’ lab-testing environment (i.e. eye tracker present) with behaviour from a less-contrived testing environment (i.e. no eye tracker present). As a result, we now know that researchers who employ eye trackers to obtain a pure measurement of where people attend with their eyes may not have been measuring as pure a sample of behaviour as they thought.  The presence of mobile technologies is an interesting example of a cognitive extension. That individuals tend to think of the internet essentially as an extension of their own mind, and as a place to store and retrieve knowledge, is well-known (Sparrow, Liu, and Wegner, 2011). Similar to the shell and bin example described above, mobile technologies are ever-present and well designed to be exploited for offloading an internal cognitive load. However, the influence of their presence in the context of completing cognitive tasks is still poorly understood. In fact, technology-related concerns have often centered on the notion that device overuse is producing long-term damage to the cognitive systems of humans. A situated cognition approach therefore introduces the possibility that being actively engaged with technological devices may not be the only relevant consideration. Despite the fact that this may operate outside of one's explicit   125 intentions and awareness, with regard to fluid intelligence the combination of an individual and their phone created a cognitive system that was either more efficient (if the individual was a high technology user) or less efficient (if the individual was a low technology user). These data drive home again the point that assessing cognitive performance without consideration for the environment (in this case, the presence of a device), would have failed to yield a complete story.  6.4   Limitations Although there are benefits to including several different literatures in a dissertation (in this case, one benefit had to do with being able to showcase the value of a situated approach to cognition across vastly different contexts), there are also drawbacks. For instance, while breadth of scope is maximized, depth can be compromised. In the current dissertation, this took the form of not having the opportunity to thoroughly explore possible mechanisms for the behaviours observed in each of the chapters. For example, although speculations were made, there were no experimental explanations for why wearing an eye tracker made individuals less willing to cheat or why the presence of a phone impacted fluid intelligence scores. Relatedly, it was outside the scope of the current thesis to follow-up on some intriguing questions, such as: why does phone presence matter for fluid intelligence but not for some other cognitive domains? In addition, it was outside the scope of the current work to take the situated cognition approach even further and employ a greater degree of task naturalness. For example, it would have been interesting to try and measure various cognitive domains in the ‘real world’ and compare those findings to the lab-based tasks. Lastly, all participants in this research were WEIRD (Western, Educated, Industrialized, Rich, and Democratic) (Henrich, Heine, & Norenzayan, 2010). Given the focus on viewing cognition as a situated phenomenon, it is unfortunate that the testing of a more diverse group was not possible.    126 6.5   Conclusions In addition to the specific questions put forth in each individual chapter, my thesis as a whole is built on one foundational question: how does a situated approach to cognition add explanatory value to research findings? Throughout the dissertation, across four data chapters and seven empirical investigations, I demonstrate how use of this framework carries important methodological implications  and has the power to uncover effects that may have otherwise been missed.127  References Aguila, B.R. (2019). "Codependency traits and the mere presence of a cell phone". EWU Masters Thesis Collection. 547. https://dc.ewu.edu/theses/547   Alzahabi, R., & Becker, M. W. (2013). The association between media multitasking, task-        switching, and dual-task performance. Journal of Experimental Psychology: Human       Perception and Performance, 39(5), 1485. Alzahabi, R., Becker, M. W., & Hambrick, D. Z. (2017). Investigating the relationship       between media multitasking and processes involved in task-switching. Journal of       Experimental Psychology: Human Perception and Performance,43(11), 1872-1894.       doi:10.1037/xhp0000412 Anderson, M., Jiang, J., Anderson, M., & Jiang, J. (2018, November 30). Teens, Social Media       & Technology 2018. Retrieved from https://www.pewinternet.org/2018/05/31/teens-social-      media-technology-2018/ Ando, A. W., & Gosselin, A. Y. (2005). Recycling in multifamily dwellings: Does       convenience matter? Economic inquiry, 43(2), 426-438. Argo, J. J., Dahl, D. W., & Manchanda, R. V. (2005). The influence of a mere social presence       in a retail context. Journal of consumer research, 32(2), 207-212. Austin, J. Hatfield, D.B., Grindle, A.C., Bailey, J.S. (1993). Increasing recycling in office       environments: The effects of specific, informative cues. Journal of Applied Behavior       Analysis, 26, 247-253. Balta, Ö. Ç., & Horzum, M. B. (2008). The factors that affect internet addiction of students in a web based learning environment. Ankara University, Journal of Faculty of Educational Sciences, 41(1), 187-205. Bateson, M., Nettle, D., & Roberts, G. (2006). Cues of being watched enhance cooperation in a real-world setting. Biology letters, 2(3), 412-414. Baer, S., Bogusz, E., & Green, D. A. (2011). Stuck on screens: Patterns of computer and      gaming station use in youth seen in a psychiatric clinic. Journal of the Canadian Academy of      Child and Adolescent Psychiatry, 20(2), 86.  128  Beard, K. W., & Wolf, E. M. (2001). Modification in the proposed diagnostic criteria for Internet addiction. Cyberpsychology & behavior, 4(3), 377-383.  Bernstad, A. (2014). Household food waste separation behavior and the importance of       convenience. Waste management, 34(7), 1317-1323. Beuckels, E., Kazakova, S., Cauberghe, V., Hudders, L., & De Pelsmacker, P. (2019). Freedom makes you lose control. European Journal of Marketing.  Bezdjian, S., Baker, L. A., Lozano, D. I., & Raine, A. (2009). Assessing inattention and       impulsivity in children during the Go/NoGo task. British Journal of Developmental       Psychology,27(2), 365-383. doi:10.1348/026151008x314919 Bindemann, M., Mike Burton, A., & Langton, S. R. (2008). How do eye gaze and facial expression interact?. Visual Cognition, 16(6), 708-733. Boot, W., Blakely, D., & Simons, D. (2011). Do action video games improve perception and cognition? Frontiers in Psychology, 2(226), 1–6. doi:10.3389/fpsyg.2011.00226 Boyd, D.& Marwick, A. E. (2014). Networked privacy: How teenagers negotiate context in                                                                                social media. New media & society, 16(7), 1051-1067. Broadbent, D. E. (1971). Decision and stress.   Brothers, K. J., Krantz, P. J., & McClannahan, L. E. (1994). Office paper recycling: A function       of container proximity. Journal of Applied Behavior Analysis, 27(1), 153-160. Brown, R. A. (2002). Evaluation of a behavioral measure of risk taking: the Balloon Analogue Buck, R., Losow, J. I., Murphy, M. M., & Costanzo, P. (1992). Social facilitation and inhibition of emotional expression and communication. Journal of Personality and Social Psychology, 63(6), 962. Cain, M. S., & Mitroff, S. R. (2011). Distractor filtering in media multitaskers. Perception,       40(10), 1183-1192. doi: 10.1068/p7017 Cao, F., Su, L., Liu, T., & Gao, X. (2007). The relationship between impulsivity and Internet       addiction in a sample of Chinese adolescents. European Psychiatry, 22(7), 466-471. Carr, N. (2011). The shallows: What the Internet is doing to our brains. WW Norton &       Company. 129  Carver, C. S., & Scheier, M. F. (1981). The self-attention-induced feedback loop and social facilitation. Journal of Experimental Social Psychology, 17(6), 545-568. Carver, C. S., & Scheier, M. F.(1981). Attention and self-regulation: A control-theory approach to human behavior.  Chaiken, S., & Trope, Y. (1999). Dual-process models in social psychology. New York: Guilford. Chartier, Y. (2014). Safe management of wastes from health-care activities. World Health       Organization..  Cheever, N. A., Rosen, L. D., Carrier, L. M., & Chavez, A. (2014). Out of sight is not out of       mind: The impact of restricting wireless mobile device use on anxiety levels among low,       moderate and high users. Computers in Human Behavior, 37, 290–297.       https://doi.org/10.1016/j.chb.2014.05.002 Chisholm, J. D., & Kingstone, A. (2015). Action video games and improved attentional       control: Disentangling selection-and response-based processes. Psychonomic bulletin &       review, 22(5), 1430-1436. Cisco, J. (2014). Teaching the literature review: A practical approach for college instructors. Teaching and Learning Inquiry, 2(2), 41-57. Clark, A (1989). Microcognition. Cambridge, MA: MIT Press Clement, J., & Miles, M. (2018). Screen schooled: Two veteran teachers expose how technology overuse is making our kids dumber. Black Inc.. Çuhadar, C. (2012). Exploration of problematic Internet use and social interaction anxiety among Turkish pre-service teachers. Computers & Education, 59(2), 173-181. Clark, A. (2012). Embodied, embedded, and extended cognition. The Cambridge handbook of cognitive science, 275-291.  Cole, C., Osmani, M., Quddus, M., Wheatley, A., & Kay, K. (2014). Towards a zero waste       strategy for an local English authority. Resources, Conservation and Recycling, 89, 64-75. Collins, E., Freeman, J., & Chamarro-Premuzic, T. (2012). Personality traits associated with       problematic and non-problematic massively multiplayer online role playing game use.       Personality and Individual Differences, 52(2), 133-138. 130  Cornblatt, B. A., Risch, N. J., Faris, G., Friedman, D., & Erlenmeyer-Kimling, L. (1988). The Continuous Performance Test, identical pairs version (CPT-IP): I. New findings about sustained attention in normal families. Psychiatry research, 26(2), 223-238.  Çuhadar, C. (2012). Exploration of problematic Internet use and social interaction anxiety       among Turkish pre-service teachers. Computers & Education,59(2), 173-181.       doi:10.1016/j.compedu.2011.12.029 Davies, J. M., & Mazumder, A. (2003). Health and environmental policy issues in Canada: the       role of watershed management in sustaining clean drinking water quality at surface sources.       Journal of Environmental Management, 68, 273-286. De Young, R., Boerschig, S., Carney, S., Dillenbeck, A., Elster, M., Horst, S., Thomson, B.       (1995). Recycling in multi-family dwellings: Increasing participation and decreasing       contamination. Population and Environment, 16(3), 253-267. Derksen, L., & Gartrell, J. (1993). The social context of recycling. American sociological review, 434-442.. Deutsche Telekom AG (2012), “Smart Payments—How the Cell Phone Becomes a Wallet,”       research report, http://www.studie-life.de/en/life-reports/smart-payments/. Domina, T., & Koch, K. (2002). Convenience and frequency of recycling: implications for       including textiles in curbside recycling programs. Environment and behavior, 34(2), 216-      238. Dong, G., Zhou, H., & Zhao, X. (2011). Male Internet addicts show impaired executive control       ability: evidence from a color-word Stroop task. Neuroscience letters, 499(2), 114-118. Duffy, S. & Verges, M. (2009). It matters a hole lot: Perceptual affordances of waste container       influence recycling compliance. Environment & Behavior, 41, 741-749. Egli, E. A., & Meyers, L. S. (1984). The role of video game playing in adolescent life: Is there       reason to be concerned?. Bulletin of the psychonomic society, 22(4), 309-312. Ehrenstein, C (2012). Neuroscience: The Internet is Destroying Your Brain Environment and Climate Change Canada (2016). Canadian Environmental Sustainability       Indicators: Greenhouse Gas Emissions. Retrieved from: http://www.ec.gc.ca/indicateurs-      indicators/default.asp?lang=en&n=FBF8455E-1 Environment and Climate Change Canada. (2014). Municipal Solid Waste and Greenhouse  131       Gases. Retrieved from: http://www.ec.gc.ca/gdd-mw/default.asp?lang=En&n=6f92e701-1 Environmental Protection Agency (2015). Waste Reduction Model. Retrieved from       http://epa.gov/epawaste/conserve/tools/warm/index.html Environmental Protection Agency. (2010). Inventory of U.S. Greenhouse Gas Emissions and       Sinks: 1990-2008. (EPA publication no. EPA 430-R-10-006.) Retrieved from:       http://www.epa.gov/climatechange/ghgemissions/usinventoryreport/archive.html Fallde, M. (2015). Can area managers connect policy and tenants? Implementation and      diffusion of a new waste management system in Linköping, Sweden. Journal of      Environmental Planning and Management, 58(5), 932-947. Ferguson, A. M., McLean, D., & Risko, E. F. (2015). Answers at your fingertips: Access to the       Internet influences willingness to answer questions. Consciousness and cognition, 37, 91-      102. Fisher, M., Goddu, M. K., & Keil, F. C. (2015). Searching for explanations: How the Internet       inflates estimates of internal knowledge. Journal of experimental psychology: General,       144(3), 674. Foulsham, T., Walker, E., & Kingstone, A. (2011). The where, what and when of gaze allocation in the lab and the natural environment. Vision research, 51(17), 1920-1931.  Fowler, H. N. (1925). Plato in Twelve Volumes, Vol. 9. Cambridge, MA, and London Friesen, C. K., & Kingstone, A. (1998). The eyes have it! Reflexive orienting is triggered by nonpredictive gaze. Psychonomic bulletin & review, 5(3), 490-495. Friesen, C. K., Moore, C., & Kingstone, A. (2005). Does gaze direction really trigger a reflexive shift of spatial attention?. Brain and cognition, 57(1), 66-69. Gallup, A. C., Chong, A., & Couzin, I. D. (2012). The directional flow of visual information transfer between pedestrians. Biology letters, 8(4), 520-522. Gamba, R. J. & Oskamp, S. (1994). Factors Influencing Community Residents’ Participation       in Commingled Curbside Recycling Programs. Environment & Behavior, 26(5), 587-612. George, M. J., & Odgers, C. L. (2015). Seven fears and the science of how mobile       technologies may be influencing adolescents in the digital age. Perspectives on       Psychological Science, 10(6), 832-851. Giedd, J. N. (2012). The digital revolution and adolescent brain evolution. Journal of  132       Adolescent Health, 51(2), 101-105. Giusti, L. (2009). A review of waste management practices and their impact on human health.       Waste Management, 29(8), 2227-2239. Gorman, T. E., & Green, C. S. (2016). Short-term mindfulness intervention reduces the       negative attentional effects associated with heavy media multitasking. Scientific       Reports,6(1). doi:10.1038/srep24542 Green, C. S., & Bavelier, D. (2012). Learning, attentional control, and action video games. Current biology, 22(6), R197-R206. Greenfield, P. M. (2009). Technology and informal education: What is taught, what is learned.       Science, 323(5910), 69-71. Greenfield, P. M. (2013). The changing psychology of culture from 1800 through 2000. Psychological science, 24(9), 1722-1731. Guerin, B. (1986). Mere presence effects in humans: A review. Journal of experimental social psychology, 22(1), 38-77. Guerin, B. (2010). Social facilitation. The Corsini encyclopedia of psychology, 1-2. Haley, K. J., & Fessler, D. M. (2005). Nobody's watching?: Subtle cues affect generosity in an anonymous economic game. Evolution and Human behavior, 26(3), 245-256.  Hallsworth, M. (2015). John List, Robert Metcalfe and Ivo Vlaev (2014),” The Behavioralist as Tax Collector,” Using Natural Field Experiments to Enhance Tax Compliance.”. NBER Working Paper, 20007. Hanks, A. S., Just, D. R., Smith, L. E., & Wansink, B. (2012). Healthy convenience: nudging s      students toward healthier choices in the lunchroom. Journal of Public Health, 34(3), 370-      376. Hossain, M., Santhanam, A., Norulaini, N., & Omar, A.(2011). Clinical solid waste       management practices and its impact on human health and environment–A review. Waste       management, 31(4), 754-766. Hottle, T. A., Bilec, M. M., Brown, N. R., Landis, A. E. (2015). Toward zero waste:       Composting and recycling for sustainable venue based events. Waste Management, 38, 86-      94. 133  İmren, M., & Tekman, H. G. (2019). The Relationship Between Media Multytasking, Working Memory and Sustained Attention. Social Sciences Review of the Faculty of Sciences & Letters University of Uludag/Fen Edebiyat Fakültesi Sosyal Bilimler Dergisi, 20(37).  Jarchow, M. E., Rice, J. W., Ritson, R. M., & Hargreaves, S. K. (2011). Awareness and        convenience are important in increasing conference sustainability. Sustainability Science,        6(2), 253-254. Kahneman, D. (2011). Thinking, fast and slow. Macmillan.   Karak, T., Bhagat, R. M., & Bhattacharyya, P. (2012). Municipal solid waste generation,       composition, and management: The world scenario. Critical Reviews in Environmental       Science and Technology, 42(15), 1509. Khalid, A., Arshad, M., Anjum, M., Mahmood, T., & Dawson, L. (2011). The anaerobic       digestion of solid organic waste. Waste Management, 31(8), 1737-1744. Khoury, J. M., Neves, M. D. C. L. D., Roque, M. A. V., Queiroz, D. A. D. B., Corrêa de Freitas, A. A., de Fátima, Â., ... & Garcia, F. D. (2019). Is there a role for cannabidiol in psychiatry?. The World Journal of Biological Psychiatry, 20(2), 101-116.  Kim, E. J., Namkoong, K., Ku, T., & Kim, S. J. (2008). The relationship between online game       addiction and aggression, self-control and narcissistic personality traits. European       Psychiatry,23(3), 212-218. doi:10.1016/j.eurpsy.2007.10.010 Kingstone, A., Smilek, D., & Eastwood, J. D. (2008). Cognitive ethology: A new approach for       studying human cognition. British Journal of Psychology, 99(3), 317-340. Kingstone, A., Smilek, D., Ristic, J., Kelland Friesen, C., & Eastwood, J. D. (2003). Attention,          researchers! It is time to take a look at the real world. Current Directions in Psychological       Science, 12(5), 176-180. Király, O., Nagygyörgy, K., Griffiths, M. D., & Demetrovics, Z. (2014). Problematic online gaming. In Behavioral addictions (pp. 61-97). Academic Press.  Ko, C. H., Hsiao, S., Liu, G. C., Yen, J. Y., Yang, M. J., & Yen, C. F. (2010). The       characteristics of decision making, potential to take risks, and personality of college       students with Internet addiction. Psychiatry research, 175(1), 121-125. Kushlev, K., Hunter, J. F., Proulx, J., Pressman, S. D., & Dunn, E. (2019). Smartphones  134       reduce smiles between strangers. Computers in Human Behavior,91, 12-16.       doi:10.1016/j.chb.2018.09.023 Kushlev, K. (2015). Digitally connected, socially disconnected: can smartphones compromise the benefits of interacting with others? (Doctoral dissertation, University of British Columbia).  Kuss, D. J., & Lopez-Fernandez, O. (2016). Internet addiction and problematic Internet use: A       systematic review of clinical research. World journal of psychiatry, 6(1), 143. Kuss, D. J., D Griffiths, M., Karila, L., & Billieux, J. (2014). Internet addiction: A systematic review of epidemiological research for the last decade. Current pharmaceutical design, 20(25), 4026-4052. Laidlaw, K. E., Foulsham, T., Kuhn, G., & Kingstone, A. (2011). Potential social interactions are important to social attention. Proceedings of the National Academy of Sciences, 108(14), 5548-5553. Lakhan, C. (2016). Out of sight, out of mind: issues and obstacles to recycling in Ontario’s       multi residential buildings. Resources Conservation and Recycling, 108:1-9. Lejuez, C. W., Read, J. P., Kahler, C. W., Richards, J. B., Ramsey, S. E., Stuart, G. L., ... & Lenhart, A., Duggan, M., Perrin, A., Stepler, R., Rainie, H., & Parker, K. (2015). Teens, social       media & technology overview 2015 (pp. 04-09). Pew Research Center [Internet &       American Life Project]. Levitin, D. (2014). The organized mind: Thinking straight in the age of information overload.       UK: Penguin. Li, Y., Park, S. Y., Zhu, J. (2011). Solid-state anaerobic digestion for methane production from       organic waste. Renewable and Sustainable Energy Reviews, 15(1), 821-826. Lin, X., Zhou, H., Dong, G., & Du, X. (2015). Impaired risk evaluation in people with Internet       gaming disorder: fMRI evidence from a probability discounting task. Progress in Neuro-      Psychopharmacology and Biological Psychiatry, 56, 142-148. Little, M., Berg, I., Luijten, M., Rooij, A. J., Keemink, L., & Franken, I. H. (2012). Error       processing and response inhibition in excessive computer game players: an even-related       potential study. Addict Biol, 17(5), 934-947. Liu, A., Ren, F., Lin, W. Y., &Wang, J.-Y. (2015). A review of municipal solid waste  135      environmental standards with a focus on incinerator residues. International Journal of      Sustainable Built Environment, 4(2), 165-188. Loh, K. K., & Kanai, R. (2014). Higher Media Multi-Tasking Activity Is Associated with       Smaller Gray-Matter Density in the Anterior Cingulate Cortex. PLoS ONE,9(9).       doi:10.1371/journal.pone.0106698  Ludwig, T. D., Gray, T. W., & Rowell, A. (1998). Increasing recycling in academic buildings:        A systematic replication. Journal of Applied Behavior Analysis, 31(4), 683-686. Lui, K. F., & Wong, A. C. N. (2012). Does media multitasking always hurt? A positive       correlation between multitasking and multisensory integration. Psychonomic bulletin &       review, 19(4), 647-653. Lui, Y. M. (2012). Advances in matrix manifolds for computer vision. Image and Vision Computing, 30(6-7), 380-388. Luijten, M., Meerkerk, G. J., Franken, I. H., van de Wetering, B. J., & Schoenmakers, T. M.       (2015). An fMRI study of cognitive control in problem gamers. Psychiatry Research:       Neuroimaging, 231(3), 262-268. Lumley, F. H., & Calhoon, S. W. (1934). Memory span for words presented auditorially. Journal of Applied Psychology, 18(6), 773.  Lyngs, U. (2017, May). 'It's More Fun With My Phone': A Replication Study of Cell Phone       Presence and Task Performance. In Proceedings of the 2017 CHI Conference Extended      Abstracts on Human Factors in Computing Systems (pp. 136-141). ACM. Maccoby, E. E. (1951). Television: Its impact on school children. Public Opinion Quarterly,       15(3), 421-444. Madden, M., Lenhart, A., Cortesi, S., Gasser, U., Duggan, M., Smith, A., & Beaton, M. (2013). Teens, social media, and privacy. Pew Research Center, 21(1055), 2-86. Magen, H. (2017). The relations between executive functions, media multitasking and       polychronicity. Computers in Human Behavior,67, 1-9. doi:10.1016/j.chb.2016.10.011 Margai, F. (1997). Analyzing changes in waste reduction behavior in a low-income urban       community following a public outreach program. Environment and Behavior, 29: 769–792. McQuaid, R. W., & Murdoch, A. R. (1996). Recycling policy in areas of low income and  136       multi-storey housing. Journal of Environmental Planning and Management, 39(4), 545-562. Mei, S., Montanari, A., & Nguyen, P. M. (2018). A mean field view of the landscape of two-layer neural networks. Proceedings of the National Academy of Sciences, 115(33), E7665-E7671.  Melia, S., Parkhurst, G., & Barton, H. (2011). The paradox of intensification.Transport Policy,       18(1), 46-52. Mills, K. L. (2016). Possible effects of internet use on cognitive development in adolescence.       Media and Communication, 4(3). Minear, M., Brasher, F., McCurdy, M., Lewis, J., & Younggren, A. (2013). Working memory,       fluid intelligence, and impulsiveness in heavy media multitaskers. Psychonomic bulletin &       review, 20(6), 1274-1281. Mirc, N., Rouzies, A., & Teerikangas, S. (2017). Do academics actually collaborate in the study of interdisciplinary phenomena? A look at half a century of research on mergers and acquisitions. European Management Review, 14(3), 333-357. Näsi, M., & Koivusilta, L. (2013). Internet and everyday life: The perceived implications of       internet use on memory and ability to concentrate. Cyberpsychology, Behavior, and Social       Networking, 16(2), 88-93. Nasiopoulos, E., Risko, E. F., Foulsham, T., & Kingstone, A. (2015). Wearable computing: Will it make people prosocial?. British Journal of Psychology, 106(2), 209-216. Nie, J., Zhang, W., Chen, J., & Li, W. (2016). Impaired inhibition and working memory in       response to internet-related words among adolescents with internet addiction: A       comparison with attention-deficit/hyperactivity disorder. Psychiatry research, 236, 28-34. Nixon, H., & Saphores, J-D. M. (2009). Information and the decision to recycle: Results from       a survey of US households. Journal of Environmental Planning and Management, 52(2),       257-277. Odacı, H., & Kalkan, M. (2010). Problematic Internet use, loneliness and dating anxiety       among young adult university students. Computers & Education,55(3), 1091-1097.       doi:10.1016/j.compedu.2010.05.006 Ophir, E., Nass, C., & Wagner, A. D. (2009). Cognitive control in media multitaskers. Proc       Natl Acad Sci U S A, 106(37), 15583-15587. doi: 10.1073/pnas.0903620106 137  Osbaldiston, R., & Schott J.P. (2012). Environmental sustainability and behavioral science:       Met-analysis of proenvironmental behavior experiments. Environment & Behavior, 44,       257-299. Owen, A. M., McMillan, K. M., Laird, A. R., & Bullmore, E. (2005). N-­‐‑back working memory paradigm: A meta-­‐‑analysis of normative functional neuroimaging studies. Human brain mapping, 25(1), 46-59.  Owens, J., Dickerson, S., & Macintosh, D. L. (2000). Demographic covariates of residential       recycling efficiency. Environment & Behavior,  32(5), 637-650. Pew Research Center (2015), “U.S. Smartphone Use in 2015,” Report, Pew Research Center,       Washington, DC Przybylski, A. K., & Weinstein, N. (2012). Can you connect with me now? How the presence       of mobile communication technology influences face-to-face conversation quality. Journal       of Social and Personal Relationships,30(3), 237-246. doi:10.1177/0265407512453827 Quitney Anderson, J., & Rainie, L. (2012). The web is dead. Ralph, B. C., & Smilek, D. (2016). Individual differences in media multitasking and       performance on the n-back. Attention, Perception, & Psychophysics,79(2), 582-592.       doi:10.3758/s13414-016-1260-y Raven, John, John C. Raven, and John Hugh Court (1998), Manual for Raven’s Progressive       Matrices and Vocabulary Scales, San Antonio, TX: Harcourt Assessment. Reitan, R. M. (1958). Validity of the Trail Making Test as an indicator of organic brain damage. Perceptual and motor skills, 8(3), 271-276. Reschovsky, J. D., & Stone, S. E. (1994). Market incentives to encourage household waste       recycling: paying for what you throw away. Journal of policy analysis and management,       13(1), 120-139. Richtel, M. (2010). Growing up digital, wired for distraction. The New York Times, 21, 1-11. Risko, E. F., & Kingstone, A. (2011). Eyes wide shut: implied social presence, eye tracking and attention. Attention, Perception, & Psychophysics, 73(2), 291-296. Risk Task (BART). Journal of Experimental Psychology: Applied, 8(2), 75. 138  Robbins, P., & Aydede, M. (2009). A short primer on situated cognition. The Cambridge handbook of situated cognition, 3-10 Rothstein, T. M. (2018). The presence of smartphones and their impact on the quality of       parent-child interactions(Unpublished master's thesis). California State University,       Stanislaus, July. Sana, F., Weston, T., & Cepeda, N. J. (2013). Laptop multitasking hinders classroom learning       for both users and nearby peers. Computers & Education, 62, 24-31. doi:       10.1016/j.compedu.2012.10.003 Schultz, P.W., Oskamp, S., &  Mainieri, T. (1995). Who recycles and when? A review of       personal and situational factors. Journal of Environmental Psychology, 18, 106-121. Seddon, A. L., Law, A. S., Adams, A., & Simmons, F. R. (2018). Exploring the relationship       between executive functions and self-reported media-multitasking in young adults. Journal       of Cognitive Psychology,30(7), 728-742. doi:10.1080/20445911.2018.1525387 Sereda, Y., & Dembitskyi, S. (2016). Validity assessment of the symptom checklist SCL-90-R and shortened versions for the general population in Ukraine. BMC psychiatry, 16(1), 1-11. Shapiro, L. (2019). Embodied cognition. Routledge Shaw, M., & Black, D. W. (2008). Internet addiction. CNS drugs, 22(5), 353-365. Small, G., & Vorgan, G. (2008). Meet your iBrain. Scientific American Mind, 19(5), 42-49. Smilek, D., Birmingham, E., Cameron, D., Bischof, W., & Kingstone, A. (2006). Cognitive ethology and exploring attention in real-world scenes. Brain research, 1080(1), 101-119. Song, Q., Li, J., Zeng, X. (2015). Minimizing the increasing solid waste through zero waste       strategy. Journal of Cleaner Production, 104, 199-210.  Soto-Faraco, S., Morein-Zamir, S., & Kingstone, A. (2005). On audiovisual spatial synergy: The fragility of the phenomenon. Perception & Psychophysics, 67(3), 444-457. Sparrow, B., Liu, J., & Wegner, D. M. (2011). Google effects on memory: Cognitive       consequences of having information at our fingertips. science, 333(6043), 776-778. Stats Canada (2013). Composting by households in Canada. Retrieved from  http://www.       statcan.gc.ca/pub/16-002-x/2013001/article/11848-eng.pdf Stevens, B.J. (1999). Multi-Family Recycling: Costs, Diversion, and Program Characteristics.  139       Washington, DC: U.S. Conference of Mayors Thornton, B., Faires, A., Robbins, M., & Rollins, E. (2014). The mere presence of a cell phone       may be distracting. Social Psychology. Tonglet, M. Phillips, P.S., Bates, M.P. (2004). Determining the drivers for household pro-      environmental behavior: Waste minimization compared to recycling. Resources,       Conservation and Recycling, 42, 27-49. Townsend, T. G., Powell, J., Jain, P., Xu, Q., Tolaymat, T., & Reinhart, D. (2015). Sustainable       Practices for Landfill Design and Operation. New York: Springer Science+Business Media       New York Tripathi, G., & Ahad, M. A. (2017). Impact of excessive use of internet on cognitive development of youngsters. International Journal of Information Technology, 9(3), 281-286. Turner, M. L., & Engle, R. W. (1989). Is working memory capacity task dependent?. Journal of memory and language, 28(2), 127-154. Twenge, J. M., & Campbell, W. K. (2019). Media use is linked to lower psychological well-being: evidence from three datasets. Psychiatric Quarterly, 90(2), 311-331.  Uncapher, M. R., Thieu, M. K., & Wagner, A. D. (2015). Media multitasking and memory:       Differences in working memory and long-term memory. Psychonomic Bulletin &       Review,23(2), 483-490. doi:10.3758/s13423-015-0907-3 Uncapher, M. R., Thieu, M. K., & Wagner, A. D. (2016). Media multitasking and memory: Differences in working memory and long-term memory. Psychonomic bulletin & review, 23(2), 483-490.  Unsworth, N., Heitz, R. P., Schrock, J. C., & Engle, R. W. (2005). An automated version of       the operation span task. Behavior Research Methods,37(3), 498-505.       doi:10.3758/bf03192720 Vining, J., & Ebreo, A. (1990). What makes a recycler? A comparison of recyclers and       nonrecyclers. Environment and behavior, 22(1), 55-73. Wagner, T. P. (2011). Compact fluorescent lights and the impact of convenience and       knowledge on household recycling rates. Waste management, 31(6), 1300-1306. Wansink, B., & Hanks, A. S. (2013). Slim by design: serving healthy foods first in buffet lines  140       improves overall meal selection. PloS one, 8(10), e77055. Ward, A. F. (2013). One with the Cloud: Why people mistake the Internet's knowledge for       their own (Doctoral dissertation). Ward, A. F., Duke, K., Gneezy, A., & Bos, M. W. (2017). Brain Drain: The Mere Presence of       One’s Own Smartphone Reduces Available Cognitive Capacity. Journal of the Association       for Consumer Research,2(2), 140-154. doi:10.1086/691462 Watson, L. (2015). Humans have shorter attention span than goldfish, thanks to smartphones. The telegraph, 15. Westergård, R. (1996). Digestion by WAASA Process of Optically Separated Waste. In The      Science of Composting (pp. 758-763). Springer Netherlands. Widyanto, L., & McMurran, M. (2004). The psychometric properties of the internet addiction test. Cyberpsychology & behavior, 7(4), 443-450. Wilmer, H. H., Sherman, L. E., & Chein, J. M. (2017). Smartphones and Cognition: A review       of research exploring the links between mobile technology habits and cognitive       functioning. Frontiers in Psychology, 8. Wilson, R., & Clark, A. (2009). A short primer on situated cognition. In The Cambridge handbook of situated cognition. New York, NY: Cambridge University Press.  Winnick, M. (2016). Putting a Finger on Our Phone Obsession. Retrieved from       https://blog.dscout.com/mobile-touches Wiradhany, W., & Nieuwenstein, M. R. (2017). Correction to: Cognitive control in media       multitaskers: Two replication studies and a meta-Analysis. Attention, Perception, &       Psychophysics,80(2), 608-608. doi:10.3758/s13414-017-1456-9 Worstall, T. (2013, March 24). More People Have Mobile Phones Than Toilets. Retrieved      May 22, 2020, from https://www.forbes.com/sites/timworstall/2013/03/23/more-people-     have-mobile-phones-than-toilets/ Wu, D. W.-L., DiGiacomo, A., Lenkic, P. J., Wong, V. K., & Kingstone, A. (2016). Being in a       “green” building elicits “greener” recycling, but not necessarily “better” recycling. PLoS       One, 11(1): e0145737. Yao, Y. W., Wang, L. J., Yip, S. W., Chen, P. R., Li, S., Xu, J., ... & Fang, X. Y. (2015).       Impaired decision-making under risk is associated with gaming-specific inhibition deficits   141       among college students with Internet gaming disorder. Psychiatry research, 229(1), 302-      309. Yap, J. Y., & Lim, S. W. H. (2013). Media multitasking predicts unitary versus splitting visual       focal attention. Journal of Cognitive Psychology, 25(7), 889-902. Yau, Y. (2012). Stakeholder Engagement in Waste Recycling in a High-Rise Setting.      Sustainable Development, 20, 115.  Young, K. S. (2009). Internet addiction test. Center for On-Line Addictions. Bradford, USA.      http://www. netaddiction. com/resources/internet_addiction_test. htm (accessed 30 Nov       2008). Zaman, A. U. (2015). A comprehensive review of the development of zero waste management:      Lessons learned and guidelines. Journal of Cleaner Production, 91, 12-25.   Zhou, Z., Yuan, G., & Yao, J. (2012). Cognitive biases toward Internet game-related pictures      and executive deficits in individuals with an Internet game addiction. PloS one, 7(11),      e48961. Zhou, Z., Zhou, H., & Zhu, H. (2016). Working memory, executive function and impulsivity      in Internet-addictive disorders: a comparison with pathological gambling. Acta      neuropsychiatrica, 28(02), 92-100.  Zomorodi, M. (2015). New Tech City: Bored and Brilliant by the Numbers - WNYC. / Available at: http://www.wnyc.org/story/bored-and-brilliant-challenge- tracker/ [accessed     [accessed April 16, 2015].                  142   Appendix A  Pre-Screen Questions 1. On average, during waking hours, how often do you check your phone? a) I don't own a phone b) Once per day or less c) 2-5 times per day d) 5-10 times per day e) About once every hour f) About once every 30minutes g) About once every 15minutes h) Almost constantly  2. Aside from time spent doing work, on average during waking hours, how much time do you spend on the Internet (either by gaming/phone/computer)? a) One hour or less per week b) One hour or less per day c) 2-3 hours per day d) 4 hours per day e) 5 hours per day f) 6 hours per day g) 7 hours per day h) more than 7 hours per day  3. I feel that my internet or smartphone use (any kind: gaming, social media, online shopping etc) is negatively impacting my life in some way a) strongly disagree b) disagree c) slightly disagree d) slightly agree e) agree f) strongly agree  4. Other people have told me that they think my internet or smartphone use (any kind: gaming, social media, online shopping etc) is negatively impacting my life in some way a) no one has ever mentioned this to me b) someone has mentioned this to me but they were joking c) someone has mentioned this to me out of concern d) more than one person has mentioned this to me out of concern   143   1Technology Usage Questionnaire  Subject ID: _____Age:  ________  Gender:        M       F     Non-binary                  Ethnicity: ________    Where were you born? ________ Is English your first language?  Yes   No     If No, how many years have you spoken English? _____       What is your first language? __________ Please do your best to estimate the following. We would like to get a sense for the kinds of technology you use and how often you use them.  Note- Do not include any technology that is used for school/ work purposes.  Type Time PER DAY. Circle one What kind? Circle as many that apply Do you use this PRIMARILY on your phone or computer?  Circle one Texting/ Messaging No time 0-30 minutes 30 mins- 1 hour 1-2 hours 2-3 hours 3-4 hours 4-5 hours 5-6 hours 6+ hours iMessage/Text Whatsapp   Other:  Phone  Comp  Equal 144  Social Media No time 0-30 minutes 30 mins- 1 hour 1-2  hours 2-3 hours 3-4 hours 4-5 hours 5-6 hours   6+ hours   Facebook Facebook Messenger Instagram Snapchat Twitter Youtube Reddit Other: ___________ Phone  Comp  Equal Gaming No time 0-30 minutes 30 mins- 1 hour 1-2 hours 2-3 hours 3-4 hours 4-5 hours 5-6 hours   6+ hours Shooter Action Role-Playing Sports Adventure Fighting Strategy Racing Other: Phone  Comp  Equal 145  Devices No time 0-30 minutes 30 mins- 1 hour 1-2 hours 2-3 hours 3-4 hours 4-5 hours 5-6 hours   6+ hours Fitbit Apple-watch Drone/ Digital Camera Technological Aid  iBook Other:  Phone  Comp  Equal Email (not for school/ work) No time 0-30 minutes 30 mins- 1 hour 1-2 hours 2-3 hours 3-4 hours 4-5 hours 5- 6 hours 6+ hours   Phone  Comp  Equal Web Browsing No time Online Forum Shopping Phone  Comp  Equal 146  0-30 minutes 30 mins- 1 hour 1-2 hours 2-3 hours 3-4 hours 4-5 hours 5-6 hours 6+ hours Googling  Other:  Videos No time 0-30 minutes 30 mins- 1 hour 1-2 hours 2-3 hours 3-4 hours 4-5 hours 5-6 hours 6+ hours Netflix Youtube Cable Movies/ TV shows Phone  Comp  Equal  147  Other?   No time 0-30 minutes 30 mins- 1 hour 1-2 hours 2-3 hours 3-4 hours 4-5 hours 5-6 hours 6+ hours Explain:  Phone  Comp  Equal            

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}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            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:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0394328/manifest

Comment

Related Items