F R E Q U E N C Y - C H A N N E L I N T E R A C T I O N S O F T H E A U D I T O R Y S T E A D Y - S T A T E R E S P O N S E S A T D I F F E R E N T L E V E L S O F T H E A U D I T O R Y P A T H W A Y S by M A X I N E T. A R M S T R O N G B.Sc. , McMaster University, 2004 A THESIS S U B M I T T E D I N P A R T I A L F U L F I L M E N T O F T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F M A S T E R O F S C I E N C E in T H E F A C U L T Y O F G R A D U A T E S T U D I E S (Audiology and Speech Sciences) T H E U N I V E R S I T Y O F B R I T I S H C O L U M B I A September 2006 © Maxine T. Armstrong, 2006 Abstract This study evaluated the effect of 3 different modulation rates and 3 different rate separations on the interactions (response amplitudes) of multiple A S S R s . Responses from 12 normal-hearing subjects were examined using three amplitude-modulation rates (14, 40 and 80 Hz) and four conditions: one tone (1000 Hz) , two tones (1000 and 2000 H z , presented monotically or dichotically) and four tones (500, 1000, 2000 and 4000 H z , presented monotically). Within the multiple-tone conditions, rate separation between the amplitude-modulated tones was varied from 2 to 6 Hz . A 1000-Hz tone served as the baseline condition and interactions between tones were measured as a function of the change in response amplitude from this baseline. Results indicate separation between modulation rates had no effect on the interactions between responses. Both modulation rate and condition had significant effects on A S S R interactions. In general, interactions became greater as the number of stimuli increased from 1 to 4 tones. However, each modulation rate had a different pattern of interactions. The response amplitudes for all modulation rates were significantly decreased in the four-tone condition. The two-tone dichotic condition amplitudes were decreased from the baseline for the 40-Hz A S S R s , but not for the 14- and 80-Hz A S S R amplitudes. Furthermore, the 40-Hz and 80-H z two-tone monotic condition response amplitudes were decreased from the baseline, but not the 14-Hz A S S R amplitudes. Results from relative efficiency calculations indicate that at high intensities (80 dB SPL) , A S S R s to multiple tones are not more efficient than A S S R s to single tones for any modulation rate range. The different patterns suggest that the 14-, 40- and 80-Hz A S S R s are generated in different areas along the auditory pathways. These results may be helpful in determining the usefulness of multiple A S S R s for diagnostic testing at high intensity levels and for testing auditory functioning at different levels in the auditory pathways. [Research supported by B C M S F , N S E R C and CIHR] i i Table of Contents Abstract i i Table of Contents i i i List of Tables v List of Figures v i List of Abbreviations v i i Acknowledgements • • • • v i i i Dedication • ix Literature Review 1 Definition and history of the auditory steady-state response 1 Auditory steady-state response 1 40-Hz A S S R 2 80-Hz A S S R 4 Low-Rate A S S R 5 Generators of the A S S R 6 A S S R studies in animals . 7 Amplitude modulation sensitivity differences within the auditory pathway . . . . . 8 Latencies of A S S R s 12 Magnetoencephalography and brain electrical source analysis studies , . . 14 Mul t ip le -ASSRs r : : . . ~ . : : v . . 15 Relative efficiency 16 Physiology of interactions 17 Using A S S R and multiple-ASSR to determine threshold 20 Measurement of the A S S R 23 E E G recordings 23 A S S R analysis 24 Subject factors and recording and analysis of the A S S R 27 State of arousal . 2 7 Effect of maturation 28 Stimulus factors and A S S R s 30 Carrier frequency 30 Stimulus intensity 31 Stimulus factors and multiple-ASSRs 32 Carrier frequency 32 Modulation rate and separation 33 i i i Stimulus intensity 34 Rational for and approach of this study 34 Introduction 37 Methods . . . . " . .* 42 Subjects . • 42 Stimuli 42 Electroencephalogram (EEG) recording 45 A S S R analysis , . 46 Data analysis and statistical analysis 47 Procedure 48 Results 49 Discussion 55 Clinical Implications 58 Conclusion 60 Bibliography 62 Appendix A : Amplitude data for all conditions for all subjects 75 Appendix B : Amplitude data for all subjects pooled over separation 78 Appendix C: Relative efficiency data for all subjects 80 Appendix D : Pilot studies .". . . . . " . . . . . . . 82 Appendix E : B R E B Ethics approval 84 iv List of Tables Table 1. Modulation Frequencies of 14-Hz Stimuli . . 43 Table 2. Modulation Frequencies of 40-Hz Stimuli 43 Table 3. Modulation Frequencies of 80-Hz Stimuli 44 Table 4. Mean and Standard Deviation Amplitudes (nV) to 1000-Hz Stimuli 51 Table 5. Summary Source Table of two-way repeated-measures A N O V A for amplitude measurements comparing effects of modulation rate and condition . 52 v List of Figures Figure 1. Diagrammatic depiction of the overlap hypothesis showing the relationship between transient evoked responses and steady-state evoked responses 11 Figure 2. Fast Fourier Transform (FFT) of a mult iple-ASSR 26 Figure 3. Fast Fourier Transforms at all conditions . 50 Figure 4. Mean A S S R amplitudes 52 Figure 5. Mean relative efficiencies 54 v i List of Abbreviations Abbreviation A S S R M l M 2 M 4 D2 E E C M E G A M F M M M F F T A B R M L R S C P B M L D IAFM N A D A N O V A C R Definition Auditory steady-state Monotic single, baseline Monotic two-tone condition Monotic four-tone condition Dichotic two-tone condition Electroencephalography Magnetoencephalography Amplitude modulation Frequency modulation M ixed modulation Fast Fourier transform Auditory brainstem response Middle latency response Slow cortical potentials Binaural masking level difference Independent amplitude and frequency modulation Number of stimuli Analog-to-digital Analysis of variance Circle radius v i i Acknowledgements I would like to thank my supervisor, Dr. David Stapells, for his inspiration, expertise, and extensive input. His guidance was exceedingly valuable and taken without nearly enough thanks. I also wish to thank my committee members, Dr. Lorienne Jenstad and Dr. Anna Van Maanen, for their expertise and positive input, as well as Dr. I. Armstrong for her unfailing ability to offer constructive criticisms under tight time constraints. Additionally, I would like to thank Jenny Hatton, Susan Small, Andrew Dimitrijevic and Elais Ponton. Without the H A P L A B group to help me when I made mistakes and when I needed last-minute subjects, I may have never gotten through to the end. A special thank you to my parents Irene and Steve, as well as my sisters Emily , Caroline and Heather. They all inspire me to work to my full potential everyday. Finally, to James for countless hours of patient listening and much thoughtful advice. This research was generously supported by the British Columbia Medical Services Foundation ( B C M S F ) , the Natural Sciences and Engineering Research Council ( N S E R C ) , and the Canadian Institutes of Health Research (CIHR). v i i i Dedication the memory of my Nana. I know how proud you would be. ix Literature Review Definition and history of the auditory steady-state response Auditory steady-state response The auditory steady-state response (ASSR) is a continuous and repetitive auditory evoked potential. It has a constant phase and amplitude over time, which reflects the rate of stimulus presentation (Picton, Stapells, Perrault, Baribeau-Bfaun, & Stuss, 1984). A S S R s are generated by a stimulus that has a fast enough rate so that the transient response to one stimulus overlaps the response to the succeeding one (Picton et al., 1984). This is in contrast to the transient auditory evoked potentials, which are elicited using stimuli presented at a slow rate, so that there is no overlap for each successive response (Picton et al., 1984). A s a result, A S S R s are best described in the frequency domain and not the time domain (as reviewed in Regan, 1989). A S S R s are elicited by a periodic signal such as a train of successive clicks or tone-pips, or a tone that has been continuously amplitude modulated. Sinusoidally amplitude-modulated ( A M ) tones are frequency-specific, containing energy only at the carrier frequency plus and minus the modulation rate (Campbell, Atkinson, Francis, & Green, 1977). A S S R s elicited by sinusoidally A M tones are the focus of the present study, and several parameters that affect A S S R s w i l l be evaluated. A S S R s can be studied using various carrier frequencies as well as various amplitude-modulation rates. Carrier frequencies may be amplitude modulated at rates from 1 to 200 Hz , and they can also be frequency modulated (FM) (for a review, see Picton, 2003). Over the years, studies have determined the effect of carrier frequency, A M rate, and F M rate on A S S R s (Aoyagi, Furuse et al., 1994; Campbell et al., 1977; Dimitrijevic, John, van Roon, & Picton, 1 2001; Dobie & Wilson, 1998; John, Dimitrijevic, & Picton, 2003; John, Dimitrijevic, van Roon, & Picton, 2001; John, Lins, Boucher, & Picton, 1998; Lev i , Folsom, & Dobie, 1993; Lins, Picton, Picton, Champagne, & Durieux-Smith, 1995; Petitot, Collet, & Durrant, 2005; Picton, Skinner, Champagne, Kellett, & Maiste, 1987; Rees, Green, & Kay, 1986; Regan, 1989; Rickards & Clark, 1984; Ross, Roberts, & Pantev, 2000; Stapells, Linden, Suffield, Hamel, & Picton, 1984). In this review, I w i l l examine the effects of stimuli presented at 14, 40 and 80 H z on A S S R s . These modulation rates have been chosen since they represent responses at three different levels of the auditory pathway, and thus, w i l l provide information on A S S R s from the level of the brainstem to the auditory cortex. 40-Hz A S S R The 40-Hz A S S R was originally identified in the 1980s, and was considered a potential tool for audiometric testing (Galambos, Makeig, & Talmachoff, 1981; Stapells, Makeig, & Galambos, 1987) (Griffiths & Chambers, 1991; Kuwada, Batra, & Maher, 1986; Lev i et al., 1993; Linden, Campbell, Hamel, & Picton, 1985; Mil ford & Birchall , 1989; Plourde, Stapells, & Picton, 1991; Stapells et al., 1984). The 40-Hz A S S R was judged to be a potentially useful audiometric tool because 40-Hz stimulation produced a response larger than any other A M rate (Galambos et al., 1981; Ross et al., 2000; Stapells et al., 1984), and has a large signal-to-noise ratio making it relatively easier to detect (Ross et al., 2000). However, it is difficult to record during sleep, and demonstrates smaller response amplitudes in infants (Aoyagi et al., 1993; Stapells, Galambos, Costello, & Makeig, 1988; Suzuki & Kobayashi, 1984; Umegaki, 1995). A s a result, this response fell out of favour as an audiological tool. 2 The 40-Hz A S S R may have been discarded prematurely. There are three purposes for which it may still be useful as an audiological measure: 1) as a means to measure hearing thresholds in adults, such as in compensation cases; 2) as an objective measure of patient arousal during anaesthesia; and, 3) as a neurophysiological monitoring tool for central auditory integrity. Plourde and colleagues (1991) found that sleep attenuated the amplitude of the 40-Hz A S S R but sedation attenuated it more. This change in the 40-Hz A S S R suggests it may make a useful measure of patient arousal during sedation (Plourde & Picton, 1990; Plourde et al., 1991). Another study found that the 40-Hz A S S R was changed in patients with intra cranial lesions and thus the 40-Hz A S S R may be a useful tool for measuring central auditory integrity (Tachisawa, 1997). Furthermore, several studies have shown that the 40-Hz A S S R is reduced in patients who have brainstem or thalamic lesions (Firsching et al., 1987; Harada, Aoyagi , Suzuki, Kiren, & Koike, 1994; Spydell, Pattee, & Goldie, 1985), but not in patients who have unilateral lesions of the temporal lobe (Firsching et al., 1987; Spydell et al., 1985), suggesting that it may be useful in determining the location of damage along the auditory pathway. Dobie and Wilson (1998) and Linden et al. (1985) found that even though the 40-Hz response is attenuated during sleep or sedation, the adult A S S R is still optimally detected at 40 Hz , even during sleep (Dobie & Wilson, 1998; Linden et al., 1985). Similarly, Jerger nd colleagues found that sleep attenuated 40-Hz responses in normal adults, but did not interfere with threshold measures for this population (Jerger, Chmiel , Frost, & Coker, 1986), or for those with hearing losses (Milford & Birchall , 1989). Furthermore, the 40-Hz A S S R s were significantly better than the 80-Hz responses for threshold measures in adults due to faster 3 recording times (Van Maanen & Stapells, 2005) and these recorded thresholds that are closer to behavioural thresholds (Petitot et al., 2005; Van Maanen & Stapells, 2005). 80-Hz A S S R The 40-Hz A S S R can be affected by state of arousal, sedation or age of the subject. However, these factors do not affect the 80-Hz A S S R s in the same way as the 40-Hz A S S R s . Several studies have indicated that recordings at A M rates of 80-110 H z have relatively good signal-to-noise ratios and stable responses in infants and unconscious adults (Aoyagi, Furuse et al., 1994; Aoyagi et al., 1993; Cohen, Rickards, & Clark, 1991; John et al., 2003; John et al., 1998; Lins & Picton, 1995). In recent years, the 80-Hz A S S R has been established as a useful measure of audiometric thresholds in normal hearing adults (Aoyagi, Kiren et al., 1994; Dimitrijevic et al., 2002; Herdman & Stapells, 2001; Lins et al., 1996; Perez-Abalo et al., 2001; Ranee, Rickards, Cohen, De V i d i , & Clark, 1995), in infants and children (Aoyagi, Kiren et al., 1994; Lins et al., 1996; Perez-Abalo et al., 2001; Ranee & Rickards, 2002; Ranee et al., 1995), in high-risk children and newborns (Luts, Desloovere, & Wouters, 2006), and in those with hearing impairment (Aoyagi, Kiren et al., 1994; Dimitrijevic et al., 2002; Han, M o , L i u , Chen, & Huang, 2006; Herdman & Stapells, 2003; Lins et al., 1996). The 80-Hz mult iple-ASSR is more efficient than singly presented A M tones in infants (Hatton & Stapells, 2006). Accurate thresholds may also be obtained using the 80-Hz A S S R in individuals with hearing loss while they are wearing their hearing aids, using sounds presented in the free-field (Picton et al., 1998). Because the 80-Hz A S S R is a more useful measure of audiometric thresholds in infants than the 40-Hz A S S R , this method has been compared with the threshold measures of the 4 auditory brainstem response ( A B R ) (Cone-Wesson, Dowel l , Tomlin, Ranee, & M i n g , 2002; Johnson & Brown, 2005). In general, the A B R and the 80-Hz A S S R both provide time-efficient estimations of hearing threshold. However, only the 80-Hz A S S R is an objective measure, with thresholds determined by statistical methods (Cone-Wesson et al., 2002; Johnson & Brown, 2005; Swanepoel, Schmulian, & Hugo, 2004). B y contrast, A B R thresholds are determined by a clinician who must judge whether a response is present. Thus, both the 80-Hz and 40-Hz A S S R s have the potential to be accurate, frequency-specific, and time-efficient audiometric tools. Low-Rate A S S R A third potential audiometric tool, is the A S S R to lower rates (-14 Hz). This response has not been studied extensively, as this response may be difficult to obtain and is sometimes not consistently recorded. These complaints may arise from the fact that it is difficult to separate from the background noise in the recording (Herdman et al., 2002; Picton et al., 1987). E E G noise levels are higher at lower frequencies, resulting in greater noise levels for recordings around 14 H z compared to those around 40 or 80 H z (Herdman et al., 2002; Picton, Dimitrijevic, John, & van Roon, 2001). However, this does not mean it is impossible to record responses at low rates. These responses have been recorded, and may provide interesting insights into the processes of the auditory cortex and higher auditory areas (Dajani & Picton, 2005; Stefanatos, 1993; Stefanatos, Foley, Grover, & Doherty, 1997; Stefanatos, G . , & Ratcliff, 1989; Wong & Stapells, 2004). Stefanatos and colleagues tested the evoked responses of children using a variable-length 1000 H z carrier tone that was frequency modulated at 4 Hz . The authors found that this recording paradigm could potentially be used to detect dysfunction in auditory 5' mechanisms responsible for perception of F M signals (Stefanatos et al., 1997). Another study found that low-rate A S S R s to F M tones may be able to identify dysfunction of auditory receptive abilities in children with Landau-Kleffner syndrome (Stefanatos, 1993). Wong and Stapells (2004) compared the brainstem and cortical binaural masking level difference ( B M L D ) using 13-and 80-Hz A M tones. The B M L D existed for the low-rate A S S R but not the high-rate A S S R , indicating a difference in binaural processing between the two responses (Wong & Stapells, 2004). These studies indicate that the low-rate A S S R may provide information about the auditory pathway that is clinically relevant, such as for higher auditory processing disorders. However, there is a scarcity of information on the low-rate A S S R . More research must be completed in this area to clarify the potential uses of this response, and to compare the low-rate A S S R with A S S R s at other rates. Generators of the ASSR A s described above, A S S R s to low, mid, and high modulation rates are different from one another. These differences are likely due to the fact that A S S R s at different modulation rates arise from different levels of the auditory pathways. The following sections wi l l provide evidence demonstrating this claim. The auditory pathway is extremely complex, and not fully described. A general overview of the pathway is essential to understanding the generators of A S S R s . The inner hair cells in the cochlea are enervated by the auditory nerve, which begin the ascending auditory pathways. The auditory nerve bifurcates into a rostral branch, which travels to the anteroventral cochlear 6 nucleus, and a caudal branch, which travels to the posteroventral and dorsal cochlear nuclei (Pickles, 1986). Information from the anteroventral cochlear nucleus travels to the superior olivary complex and the subsequently to the inferior colliculus. Information from the dorsal cochlear nucleus bypasses the superior olivary complex and instead ends in the nuclei of the lateral lemniscus and inferior colliculus (Pickles, 1986). The main receiving station for the ascending pathway information from the superior olivary complex is the inferior colliculus, both ipsilaterally and contralaterally (Pickles, 1986). More specifically, the lateral nucleus of the superior olive projects bilaterally to the inferior colliculus, and the medial nucleus of the superior olive projects ipsilaterally (Pickles, 1986). From here, information travels to the medial geniculate body of the thalamus via ipsilateral and contralateral pathways. The inferior colliculus is an obligatory relay for almost all auditory input to the medial geniculate body, where it then travels to the auditory cortex in the temporal lobe. The ventral division of the medial geniculate travels directly to the primary auditory cortex, whereas the dorsal division projects to secondary auditory areas in the cortex (Pickles, 1986). Auditory information is processed in multiple parallel frequency channels, and a cochleotopic map is preserved all along the auditory pathway at least to the level of the primary auditory cortex (Pickles, 1986). A S S R studies in animals Studies using animal models collect information directly from single cells or groups of cells in the auditory pathways, thus providing insight into A S S R s along the auditory pathways that cannot be provided using human models. Animal studies have shown that the range of optimal modulation rates that single cells best respond to decreases from the auditory periphery 7 to the cortex (Palmer, 1995). Not only is the cortex more active during low-rate stimulation, the preferred rates of cells at the level of the auditory cortex fall below 20 H z (Palmer, 1995; Schreiner & Urbas, 1986). Cells in the cat cortex have largest responses at modulation frequencies below 20 Hz , and, in rabbits, responses to higher modulation frequencies reflected contributions from lower auditory structures (Kuwada et al., 2002; Schreiner & Urbas, 1986). Suzuki examined phase changes at the ipsilateral cochlear nucleus using electrodes located at different places in the cat cochlear nucleus using 80-Hz A M tones. The results demonstrated rapid phase changes at the level of the cochlear nucleus, suggesting that this structure contributes to the 80-Hz A S S R (Suzuki, 2002). Kuwada and colleagues recorded A S S R s from both the surface of the scalp and at several places in the auditory pathway, in unanesthetized rabbits. Far-field (scalp-recorded) responses are complex, involving more than one generator (Kuwada et al., 2002). However, even though multiple generators contribute to all A S S R s , they do not all contribute equally. The near-field (single unit) responses showed that low modulation rates have more contributions from cortical sources, whereas high modulation rates have more contribution from brainstem sources (Kuwada et al., 2002). Szalda and Burkard found no decrease in A S S R amplitude in the inferior colliculus for frequencies below 90 H z in anaesthetised chinchillas, but for rates at or above 90 Hz , there were amplitude decreases (Szalda & Burkard, 2005). B y contrast, a decrease in amplitude was seen for all modulation rates at the level of the auditory cortex (Szalda & Burkard, 2005). Low-rate A S S R s reflect greater cortical activity and high-rate A S S R s reflect greater subcortical activity. 8 Amplitude modulation sensitivity differences within the auditory pathway In general, all neurons in the auditory pathways can discharge at a rate that reflects an A M stimulus, and both brainstem and cortical sources are active in all A S S R s (Picton, John, Dimitrijevic, & Purcell, 2003). However, neuronal response depends on the modulation rate of stimulation. In humans, the brainstem is the primary source of A S S R s above a 50-Hz amplitude modulation rate, demonstrated by a larger neuronal response in the brainstem than any where else along the auditory pathways for those modulation rates (Mauer & Doring, 1999). Another study found that cortical sources are more active than brainstem sources in response to slower rates (12, 39 Hz) , when compared to faster rates (88 Hz) (Herdman et al., 2002). Responses to the 12-Hz stimulus were small, but were still suggestive of activation in both the brainstem and the cortex (Herdman et al., 2002). The 40-Hz A S S R closely resembles the middle latency response ( M L R ) when M L R waves are superimposed (Galambos et al., 1981). When stimuli are presented at a slower rate (e.g., 10 Hz) the M L R is completed before the next stimulus can begin. But at rates around 40 Hz , the M L R waves overlap in such a way that the superimposition of the waves eventually looks like the 40-Hz A S S R (for a review see Plourde, 2006). This is the "overlap hypothesis" (see Figure 1), and has been tested in adults and infants by Stapells et al. (1988); in adults only (Makela & Hari , 1987; Plourde et al., 1991; Santarelli et al., 1995), and in sleeping adults (Suzuki, Kobayashi, & Umegaki, 1994). For adults, the 40-Hz A S S R can be predicted reasonably well using overlapping M L R waves (Makela & Hari , 1987; Plourde et al., 1991; Stapells et al., 1988); but neither the M L R nor the 40-Hz A S S R could be obtained in infants (Stapells et al., 1988); and prediction during sleep in adults was not accurate (Suzuki et al., 9 1994). Santarelli et al. confirmed the above findings but also showed that the 40-Hz A S S R has additional periodic activity compared to the overlapped M L R s . These results suggest that the 40-H z A S S R s and M L R share at least some of the same generators, but the 40-Hz A S S R has contributions from other sources as well (Santarelli et al., 1995). Cohen found that the polarity reversal of the M L R was approximately at the level of the Sylvian fissure, in the primary auditory cortex (Cohen, 1982). It is possible that because the M L R and 40-Hz A S S R share some generators, part of the 40-Hz A S S R arises from the primary auditory cortex. Wong and Stapells compared brainstem (>60 Hz) and cortical (<60 Hz) A S S R s using the binaural masking level difference ( B M L D ) . The B M L D is exhibited by presenting an auditory stimulus and a masker to both ears. The B M L D displays the benefit of listening binaurally, and is thought to reflect brainstem processes (Wong & Stapells, 2004). Only 7-13 H z A S S R s demonstrated B M L D s , whereas 80 H z A S S R s did not; indicating different sources for A S S R s at different modulation rates, and adding weight to the conclusion that the lower rate A S S R s have contributions from areas beyond those of the 80-Hz generators (Wong & Stapells, 2004). 10 Figure 1. Diagrammatic depiction of the overlap hypothesis showing the relationship between transient evoked responses and steady-state evoked responses. The transient brainstem responses (left panel) have a single waveform (wave V - V ) occurring within 12.5 ms. Their overlap at 80 H z does not result in a larger response. The transient middle latency responses (right panel) show several waveforms (waves V - N a , Pa-Nb, Pb-Nc) each approximately 25 ms apart. Their overlap at 40 Hz results in a larger response. Wi th permission from Andrew Dimitrijevic (2006). 11 Stapells et al. (2005) suggest that the 80-Hz A S S R and the A B R may share common generators because their frequency specificity is similar, and their thresholds in dB S P L peak-to-peak equivalent are the same. The A B R originates from brainstem structures, hence, the 80-Hz A S S R may be considered the "brainstem A S S R " . In summary, A S S R s to low modulation rates (<50 Hz) have been linked to sources at the cortical level, and A S S R s to higher-rate tones (>60 Hz) have been linked to sources lower than the cortical level. Latencies of A S S R s The latency of an evoked potential can also suggest where in the auditory pathways an evoked potential arises. A longer latency represents a generator that is higher in the auditory pathway than a response with a shorter latency. A S S R s to continuous stimuli, however, do not have latencies that can be measured directly from a continuous response; nevertheless, latencies can be estimated using apparent latency. Apparent latency is an approximation of latency using an extrapolation of the relationship between the phase of a response and stimulus rate based on several closely-spaced A M rates (Picton et al., 2003). A simple calculation is used to convert phase measurements to latency measurements for several different modulation rates. This phase-versus-rate plot must be linear to be useful as a measure of apparent latency (as reviewed in Regan, 1989). Apparent latency is sensitive to a number of confounding factors; therefore, it is important to be cautious when using apparent latency, and not to confuse the measure with the absolute latency of a response (Picton et al., 2003). 12 The apparent latency of the A S S R is affected by the modulation frequency, the carrier frequency and the intensity of a tone. Modulation rate and apparent latency are inversely related (Rickards & Clark, 1984; Stapells et al., 1984); Rickards et al. (1984) state that modulation rate is the main factor contributing to A S S R apparent latency. A n inverse relationship also exists between apparent latency of A S S R s and carrier frequency (John & Picton, 2000; Picton et al., 2003). This phenomenon can be attributed to the travelling wave in the cochlea. Lower carrier frequencies require more time to travel to their place on the tonotopic map on the cochlea, thus increasing the time it takes for transduction of that sound. Finally, apparent latency of the A S S R is inversely related to stimulus intensity. John et al. noted that for the 80-Hz A S S R , as intensity increased from 35 to 75 dB S P L , the apparent latency decreased 2.4 ms (John & Picton, 2000). Rickards and Clark found that there were three distinct groups of apparent latencies: low rates (<20 Hz) , mid-rates (20-40 Hz) and high rates (>60 Hz) and they postulated that each group corresponded to the latencies of several transient evoked potentials. They suggested the low-rate A S S R s correspond to slow cortical potentials (SCPs), with apparent latencies of >60 ms; the mid-rate A S S R s correspond to middle latency responses (MLRs) , with apparent latencies of 20-50 ms; and the high-rate A S S R s correspond to brainstem responses, with latencies of <20 ms (Rickards & Clark, 1984). Apparent latency for the M L R s was also shown by Stapells et al. (1984), who obtained a mean apparent latency of 34 ms to modulation rates between 30 and 60 Hz , and Picton and colleagues (1987) who found an apparent latency of 37.2 ms for the 40-Hz A S S R . This latency is similar to the range of latencies found for the M L R . Kuwada and colleagues (1986) found an average apparent latency of 31 ms to modulation rates in the 40-Hz range, and apparent latencies of 7-9 ms to modulation rates greater than 100 H z . To explain 13 these three distinct groups of apparent latencies, Regan (1989) hypothesised the existence of separate generators within the auditory pathway. Magnetoencephalography and brain electrical source analysis studies Magnetoencephalography ( M E G ) is a measurement of the magnetic fields produced by electrical activity in the brain that, like E E G , infers activity of neurons in the brain in real time. However, M E G provides better spatial resolution than E E G , allowing localization of responses to certain areas of the brain, for example in cortical sources (Williamson, Karron, & Kaufman, 1991). M E G cannot always resolve sources deep in the brain, such as in the brainstem (Williamson et al., 1991). M E G localised the 40-Hz A S S R can be localized to the Sylvian fissure in the primary auditory cortex (Romani, Will iamson, Kaufman, & Brenner, 1982; Ross, Draganova, Picton, & Pantev, 2003). Mauer and Dbring (1999) used brain electrical source analysis to localize the generators of the A S S R s for A M rates between 24-120 Hz . The results indicate that brainstem generators are active for both low- and high-rate A S S R s , but are dominant for high-rate (>56 Hz) A S S R s , whereas cortical sources contribute to both low- and high-rate A S S R s , but are dominant for low-rate (<56 Hz) A S S R s (Mauer & Doring, 1999). Similarly, Herdman et al. (2002) found that the 39-Hz A S S R had generators in both the brainstem and cortex, and the 80-Hz A S S R had generators primarily in the brainstem. hi conclusion, low-rate (<50 Hz) and high-rate (>50 Hz) A S S R s share generators, but the dominant source is different for each response. Although both the brainstem and cortex are activated in all A S S R s , there are clear differences in the locations of A S S R s of different rates 14 along the auditory pathways. These differences are shown in animal studies and human studies that include single-cell recordings, apparent latencies, sensitivity differences along the auditory pathways, and M E G and source analysis. Multiple-ASSRs A S S R s can serve as a useful audiological technique because they can be recorded in response to multiple stimuli simultaneously (Lins & Picton, 1995). Regan was first to record responses to multiple visual stimuli by ensuring that each stimulus is modulated at a different rate (as reviewed in Regan, 1989). John and colleagues assessed several factors that affect multiple-A S S R recordings by varying the separation of the carrier frequencies of two tones, four tones, and eight tones, varying the intensity at which the tones are presented, and varying the modulation rates of each of the frequencies in the 80-Hz range (John et al., 1998; see also John et al., 2001; John, Purcell, Dimitrijevic, & Picton, 2002; Lins & Picton, 1995). Viable responses can be obtained from up to eight frequencies (four presented in each ear), as long as carrier frequencies have a separation of at least an octave and the A M rates are different for each stimulus. This technique is called mult iple-ASSR (Lins et al., 1995). The carrier frequencies can also be frequency modulated (FM) at different rates, or A M and F M , in order to produce the largest response amplitudes possible (John et al., 2001). Another combination of modulations has been to use independent amplitude and frequency modulation ( IAFM) , to determine whether they would provide a more effective way of testing multiple frequencies (Dimitrijevic, John, & Picton, 2004; Dimitrijevic et al., 2001). 15 Most studies have focussed on the 80-Hz A S S R for two reasons: 1) because the reliability of the response is good in infants and sleeping adults, and 2) preliminary research has indicated that the 40-Hz A S S R to multiply presented stimuli were no more efficient than the responses to singly presented stimuli (John et al., 1998). However, new evidence is emerging showing that mult iple-ASSR may be feasible in the 40-Hz range (Fontaine and Stapells, in preparation; Van Maanen and Stapells, 2005); and in fact, may be a more efficient measure of multiple-ASSRs in adults (Fontaine and Stapells, in preparation; Van Maanen and Stapells, 2005; Petitot et al., 2005). Few studies have directly compared the 40- and 80-Hz A S S R (Petitot et al., 2005; Van Maanen & Stapells, 2005); and no studies have examined the effects of multiple stimuli on the low-rate A S S R . Most importantly, there are no studies that compare A S S R s at all three modulation rates. Relative efficiency There is a limitation to the mult iple-ASSR technique that is present at all A M rates: response interactions. Whenever two or more stimuli are presented together, interactions occur between responses to these stimuli along the auditory pathway. These interactions cause a decrease in the amplitude of the responses compared to the response had each stimulus been presented alone (Lins et al., 1995). Such decreases in the amplitude may increase the time needed to obtain recordings, because more trials are then needed to improve the signal-to-noise ratio. These interactions are increased by a number of factors: 1) carrier frequencies that are closer together than one octave, 2) increased intensity levels, and 3) responses in the 30-50 H z range (John et al., 1998). 16 However, interactions should not preclude the use of mult iple-ASSR as a potentially useful audiometric tool. The increase in speed as a result of testing multiple frequencies can be weighed against the decrease in amplitude as a result of interactions; this is called relative efficiency. Relative efficiency can be described as follows: background noise in an E E G recording decreases at the rate of the root number of sweeps averaged (John, Purcell, Dimitrijevic, & Picton, 2002). Using multiple simultaneous stimuli rather than a single stimulus w i l l increase the amount of information available for a given recording time. However, the decrease in noise is consistent regardless of the number of stimuli being presented. Therefore, for multiple stimuli to be more efficient than a single stimulus, the decrease in amplitude that occurs when adding multiple frequencies together must not be more than 1A/N, where N is the number of stimuli (John et al., 1998). Physiology of interactions Interactions occur when multiple stimuli are presented monaurally or binaurally for both 80- and 40-Hz A S S R s (Fontaine & Stapells, in preparation; John et al., 1998). John et al. (1998) found that although interactions attenuated all tones, A M tones with higher carrier frequencies had a greater attenuating effect on A M tones with lower carrier frequencies. These results are unexpected based on the literature on the spread of masking (for review see Moore, 1995). This phenomenon has been observed in humans (John et al., 1998) as well as in animals (Dolphin & Mountain, 1993). Dolphin and Mountain (1993) assert that these interactions are similar to what is happening as a result of two-tone suppression that takes place in the cochlea. However, two-17 tone suppression does not account for all of the interactions that are occurring in multiple-A S S R s . John and colleagues (1998) hypothesized that the differences between the 40- and 80-Hz interactions that are not accounted for by two-tone suppression may be a result of interactions at the level of brainstem and/or the cortex. They studied interactions in the amplitude rate ranges of 40 and 80 H z and found first that a 2000-Hz tone amplitude-modulated at 96.9 H z significantly decreased the amplitude of a 1000-Hz tone amplitude-modulated at 80.9 H z (John et al., 1998). Secondly, they found that multiple, monaural presentation yielded significant interactions (i.e., a decrease in response amplitude) when four and eight tones amplitude-modulated in the 80-Hz range were presented simultaneously, and each carrier frequency was separated by one half octave (all A M rates were separated by 4 Hz) . Third, they found that in the 80-Hz range, at 75 dB S P L , both 2-tone and 4-tone conditions had significant interactions; however, at 35 dB S P L these interactions did not occur. Fourth, they found no increase in interactions with decreasing separation of A M rates between a 1000- and 2000-Hz tone that differed by 7.9, 5.3, 2.6, and 1.3 Hz . Finally, they found significant interactions for the 4-tone and 8-tone presentations for the 40-Hz A S S R s for stimuli presented at 60 dB S P L , but no significant interactions when two tones of 1000 and 2000 H z were presented together. More interactions occurred for the 40- than the 80-H z A S S R . John et al. (1998) concluded that multiple stimuli generated more interactions at the level of the cortex (40-Hz A S S R ) than at the level of the brainstem (80-Hz A S S R ) . Inconsistent with John et al. (1998; see final point above), using M E G Ross and colleagues (2003) found significant interactions between 1000- and 2000-Hz tones modulated in the 40-Hz range . Tones were presented in pairs using all possible combinations of 250, 500, 1 8 1000, 2000, and 4000 Hz , with modulation rates of the two tones constant at 39 and 41 Hz . The difference in findings may be a result of intensity level, as the Ross et al. (2003) study used levels of approximately 70-80 dB S P L , whereas John et al. (1998) presented the tones at 60 dB S P L . Ross et al. (2003) concluded that the characteristics of interactions of the 40-Hz A S S R are different from those at 80 Hz . However, these studies alone do not provide a comprehensive view of the 40-Hz A S S R . Several studies indicate that multiple amplitude-modulated tones in the 40-Hz range may be feasible as a threshold measure (Fontaine & Stapells, in preparation; Petitot et al., 2005; Van Maanen & Stapells, 2005). Fontaine and Stapells studied the interactions of 40-Hz A S S R s by presenting a single baseline tone, four stimuli monaurally, and eight stimuli dichotically at three different intensity levels: 30, 55, and 80 dB H L . In general, results indicated that response amplitudes decreased as the number of stimuli increased from one to eight. Despite the increased interactions for multiple stimuli conditions, multiple-stimuli conditions were more efficient than the single-stimulus condition at intensities of 30 and 55 dB H L (Fontaine & Stapells, in preparation). A t an intensity of 80 dB H L , the multiple-stimuli conditions were not more efficient than the single-stimulus condition (Fontaine & Stapells, in preparation). Van Maanen and Stapells (2005) found that both 40- and 80-Hz stimuli were accurate predictors of behavioural audiometric thresholds at all of the frequencies tested, but the 40-Hz A S S R was faster than the 80-Hz A S S R , and 40-Hz thresholds were closer to behavioural thresholds than those at 80 Hz . They used four carrier frequencies presented at the same time in one ear, each at different modulation rates, to assess hearing thresholds of normal-hearing and hearing-impaired individuals. 19 No studies have investigated A S S R interactions for low-rate stimuli (<20 Hz) , possibly due to reported difficulty in obtaining consistent low-rate responses (Herdman et al., 2002), although some success in recording has been reported (Dajani & Picton, 2005; Stefanatos, 1993; Stefanatos et al., 1989; Wong & Stapells, 2004). Further research is therefore warranted for low-rate A S S R in order to obtain more information on interactions at all levels of the auditory pathways. In summary, interactions are affected by several factors, including; modulation rate range, carrier frequency separation, intensity, and number of carrier frequencies. These factors affect the 40-Hz and the 80-Hz A S S R differently, although all differences in interactions for these A S S R s have not been determined. To date, there are no data on the interactions of A S S R s for low-rate stimuli (<20 Hz) . . Using A S S R and mult iple-ASSR to determine threshold Often, objective measures of hearing loss are necessary for a complete and accurate audiological test battery. A n "objective" measure of hearing thresholds does not require a subject response or a subjective judgement of a clinician to determine response presence (Van Maanen & Stapells, 2005). Objective tests are necessary for patients who are too young or unable to respond to behavioural testing, for patients who are suspected of functional hearing loss (hearing loss with a non-organic basis), or for the assessment of patients in medico-legal or compensation cases. Some electrophysiological measures of hearing, such as the auditory brainstem response ( A B R ) and the slow cortical potentials (SCP), require subjective judgement by the clinician in order to determine response presence or absence. A B R and SCPs are transient auditory evoked responses 20 that are currently the standard for evoked potential threshold testing in infant and adult cases, respectively (Hyde, 1994; Stapells, 2002). Although these tests provide accurate measures of hearing thresholds, they rely on the expertise and training of the clinician conducting the test (Don, Elberling, & Waring, 1984; Elberling & Don, 1984; Van Maanen & Stapells, 2005). A n objective test that does not require subjective judgement would be advantageous over the A B R or SCP . The 40-Hz and 80-Hz A S S R s provide objective measures of hearing. They can be used to assess audiological thresholds in adults, infants and children (Aoyagi, Furuse et al., 1994; Dimitrijevic et al., 2002; Griffiths & Chambers, 1991; Herdman & Stapells, 2001, 2003; Lins et al., 1996; Mi l ford & Birchall , 1989; Perez-Abalo et al., 2001; Picton et al., 1998; Ranee & Rickards, 2002; Ranee et al., 1995; Swanepoel et al., 2004; Van Maanen & Stapells, 2005) and they do not require the subjective judgement of a clinician. The carrier frequencies used in many experiments are those regularly tested in pure-tone behavioural audiometry; usually 500 through 4000 Hz . Studies comparing the 80-Hz mult iple-ASSR to the A B R have found that this method is as good as the A B R , even in people with hearing loss and in children (Cone-Wesson et al., 2002; Johnson & Brown, 2005; Vander Werff, Brown, Gienapp, & Schmidt Clay, 2002). Studies comparing the 40- and 80-Hz A S S R with the S C P found that A S S R results are as effective as the S C P for threshold testing of adults with hearing losses (Van Maanen & Stapells, 2005). However, A S S R s have the advantage over A B R and S C P of not requiring the subjective judgement of the clinician. Estimating an individual's thresholds can be completed using modulated tones presented singly or multiply. Presenting multiple stimuli can make the process more efficient (Herdman 21 and Stapells, 2001; John et al., 1998; Fontaine and Stapells, in preparation), and, in general, it outperforms the single-frequency technique (Luts & Wouters, 2005). There are three methods that can be used to determine threshold using A S S R s (Picton et al., 2003). The first method, the only one currently used clinically, is threshold bracketing. This method requires that testing begin at above-threshold and/or below-threshold intensities, then intensity is increased and decreased in specified amounts by bracketing up and down until a threshold is reached. The second method is extrapolation, where normative data is used to determine the threshold responses from above-threshold responses of the individual. The amplitude or phase of two responses are determined at two different intensities, and the slope from the difference in amplitudes is used to predict what the individual's thresholds might be at intensities other than the ones tested (Picton et al., 2003). This technique is not yet used clinically. The third approach is called the intensity sweep technique. This technique presents sounds that are below threshold and then slowly increases them until they are above the subjects' threshold. The subjects' threshold is determined by the level where the response appears/disappears (Picton et al., 2003). This technique is also not yet used clinically. When determining threshold using A S S R s , noise in the E E G is an important factor to consider. When seeking an individuals threshold, the amplitude of the response w i l l decrease as the intensity of the stimulus is decreased, until eventually there is no response (Picton et al., 2003). When looking for a response, one is always comparing the amplitude of the response to the surrounding E E G noise. If the background noise in a recording is larger than the response, the response wi l l be obscured and missed. In order to determine the difference between a real "no response", where the subject is no longer responding to the stimulus and a missed response, 22 where the response is obscured by the E E G noise, noise criteria must be employed. This requires that background noise is decreased, through more averaging, until a criterion noise level is reached. If the noise criterion is reached and the response is not significantly larger than the background noise, it is considered a "no response". More-strict noise criteria allow for more accuracy in recording thresholds; however, it is a trade-off because this increases the time it takes to test each intensity (John et al., 2002; Picton, Linden, Hamel, & Maru, 1983). Measurement of the ASSR E E G recordings A S S R s normally use a minimum of three electrodes, and their placement is important for the recording. For threshold determination, Lins et al. (1996) state that the E E G should be recorded with the non-inverting electrode at the vertex, the inverting electrode at the nape of the neck, and the ground electrode on the forehead. This placement is advised for single-channel recordings of dichotic multiply presented stimuli; and has been used extensively (e.g., Herdman & Stapells, 2001, 2003; Lins et al., 1996; Picton, 2003). The inverting electrode may also be placed on the ipsilateral mastoid, instead of the nape of the neck (Cone-Wesson et al., 2002; Ranee & Rickards, 2002; Vander Werff et al., 2002). Responses can be recorded using four electrodes with one each on both mastoid, i f the nape of the neck is not being used when recording responses from both ears. Bone-conduction A S S R s can be obtained from both the ipsilateral and contralateral mastoids of the test ear. These responses can then be compared (Small & Stapells, 2005b). Bone-conduction A S S R studies have shown that ipsilateral responses 23 are different from contralateral responses in infants; however, they are not different in adults (Small & Stapells, 2005b). The electrode configuration that provides the best signal-to-noise ratio for the high-rate A S S R (90 Hz) for infants was obtained using electrodes at the mastoid and C z (van der Reijden, Mens, & Snik, 2005). Contrary to findings in adults, this montage provided better signal-to-noise ratios than electrodes at the nape of the neck and C z (van der Reijden et al., 2005). Another study examined the A S S R s to air-conducted signals in the contralateral and ipsilateral recording channels to air-conducted stimuli. Results from this study indicate that infants' contralaterally recorded A S S R s were smaller than the ipsilateral A S S R s (Stapells & Van Maanen, 2006). Before the E E G is digitized for analysis, it is bandpass filtered with different filters depending on the modulation rate being studied [e.g., a bandpass filter of 30-250 H z for 80 H z (Herdman & Stapells, 2001); a bandpass filter of 5-100 H z for 40 H z (Van Maanen & Stapells, 2005); and a bandpass filter of 1-30 H z for 14 H z (Wong & Stapells, 2004)], and sampled by an analog-to-digital (AD) converter. The M A S T E R technique uses E E G broken down into sections that are approximately 1.02 seconds long, and 16 sections are combined to create a sweep that is approximately 16 seconds long (John et al., 1998; Lins et al., 1996; Picton et al., 1998). The rate of each stimulus is modified so that a precise integer of the A M cycles w i l l fit into a sweep. The sweeps are averaged over time. Lins et al. (1996) found that a significant response can be found after averaging between 16-64 sweeps, as determined by a specified signal-to-noise ratio. Recording the responses to multiple A M tones does not differ significantly from the above description. 24 A S S R analysis Analysis of A S S R s , which are periodic, is done either using a Fourier analyzer (Stapells et al., 1984) or a Fourier transform (Rickards & Clark, 1984), both of which yield objective measures of response amplitude and phase. Both techniques break down the E E G signal into individual frequency bins, which can be sorted using a power spectrum and phase (for example, see Figure 2), thus allowing the amplitude of the response to be observed at each frequency. Response presence can be determined objectively using statistical methods (e.g., F -Statistic, phase coherence) (Picton, John, Dimitrijevic, & Purcell, 2003). A response is deemed to be present i f the response bin has an amplitude that is significantly larger than the surrounding bins. One statistical method used to measure this is the Analysis of Variance ( A N O V A ) , where the power at the signal frequency bin is compared to the surrounding frequency bins, which yields an F-statistic (Lins et al., 1996). Figure 2 is an example of the F-statistic, where E E G frequency is represented along the x-axis and amplitude is represented along the y-axis. Responses to each of four tones modulated at approximately 80 H z are identified with triangles. Although not shown in this figure, the variability of the A S S R phase ("phase coherence") can also be used as a measure of response presence or absence (Stapells et al., 1987). The above statistical measures also apply to mult iple-ASSR recordings. 25 70 80 90 100 110 EEG FREQUENCY (Hz) Figure 2. Fast Fourier Transform (FFT) of mult iple-ASSR in the modulation rate range of 80 Hz , with carrier frequencies of 500, 1000, 2000 and 4000 H z . E E G frequency is represented along the x-axis and amplitude is represented along the y-axis. Responses to each of the four amplitude-modulated tones occur at the modulation rates and are identified with triangles. 26 The phase of an A S S R is a particular point in the time of a cycle; measured from some arbitrary zero and expressed as an angle. However, because the A S S R is a continuous response, phase is circular. Phase delay of the A S S R (in milliseconds), calculated from response phase (in degrees), is used to estimate the interval from stimulus presentation to A S S R . This creates uncertainty about the number of cycles included in the phase (i.e., the phase is more than 360°) (Picton et al., 2003). John and Picton (2000) addressed the problem by estimating the number of cycles of the stimulus that occurred before the response. In spite of this, caution must still be used when determining phase delays from the phase of the A S S R . Subject factors and recording and analysis of the ASSR State of arousal The background E E G noise for the 40- and 80-Hz A S S R s , and possibly the 14-Hz A S S R , decreases during sleep, improving the signal-to-noise ratio; but for some of these responses a corresponding decrease in signal amplitude also occurs during sleep potentially cancelling any signal-to-noise ratio benefits (Dobie & Wilson, 1998; Jerger et al., 1986; Lins et al., 1996; Pethe, Muhler, Siewert, & von Specht, 2004). The 40-Hz and 80-Hz A S S R s are differentially affected by state of arousal. The 40-Hz response is decreased during natural sleep (Dobie & Wilson, 1998; Linden et al., 1985; Lins et al., 1996; Picton et al., 1987), and further decreased by sedation (Plourde et al., 1991). The 40-H z A S S R thresholds for a 500-Hz tone have been shown to increase during sleep (Umegaki, 1995). Although it may not preclude the use of the 40-Hz A S S R as an audiological tool, this is a disadvantage, because the effect of arousal in adults is not understood completely (Picton et al., 27 2003). However, there are studies that find the 40-Hz A S S R more efficient than the A S S R s at other rates, even during sleep, due to its consistently large amplitude in adults (Dobie & Wilson, 1998; Jerger et al., 1986). A decrease in amplitude during sleep does not occur for the 80-Hz A S S R (Aoyagi, Furuse et al., 1994; Aoyagi et al., 1993; Cohen et al., 1991; Dobie & Wilson, 1998; Lev i et al., 1993; Umegaki, 1995). This is due to the fact that the response arises primarily from the brainstem, which is not as affected by arousal as generators higher in the auditory pathways (Cohen et al., 1991; Lins & Picton, 1995). Consequently, there is an improvement in the signal-to-noise ratio for the 80-Hz A S S R during sleep (Levi et al., 1993). The effect of arousal on the low-rate A S S R is unknown; however, one could predict a reduction in response amplitude as a result of sleep or sedation, because low-rate A S S R s are considered primarily a cortical response (Herdman et al., 2002). Slow cortical potentials have a reduced N I component with reduced alertness, but an increased N 2 component (Stapells, 2002). For example, N I stimuli evoked by unattended stimuli is greater with a higher level of alertness, and this response has been shown to undergo large changes as a result of sleep (Naatanen, 1992; Picton, Hillyard, & Galambos, 1976; for review, see Stapells, 2002). However, the relationship between slow transient responses and arousal is not clear. It could be inferred from transient response results that decreased alertness may also decrease the low-rate A S S R amplitude, because these two responses are thought to share at least some generators (see generators section). Further research is needed to assess the relationship between arousal and amplitude of the A S S R for both the 40-H z and low-rate (14 Hz) A S S R s . 28 Effect of maturation Maturation influences A S S R s : for example, adults, but not infants, show a large increase in amplitude of the A S S R at rates of 40 H z compared to other rates (Stapells et al., 1988; Suzuki & Kobayashi, 1984). Attempts to estimate thresholds using the 40-Hz A S S R in infants have proven to be unreliable (Stapells et al., 1988; Suzuki & Kobayashi, 1984). Stapells and colleagues used brief tone bursts to examine A S S R s in both adults and children during sleep. Results indicated that infants had a larger response in the low rates (<20 Hz), that decreased for rates up to 60 H z (Stapells et al., 1988). Suzuki and Kobayashi presented modulated tones at rates between 10-50 H z to adults and children and, like Stapells and colleagues, found that the amplitude of the infant A S S R decreased with increasing A M rates from 20-50 H z (Suzuki & Kobayashi, 1984). In contrast to the 40-Hz A S S R , the amplitude of the higher rate A S S R s (70-110 Hz) seem to be less affected by subject age (Aoyagi et al., 1993; Lev i et al., 1993). In fact, Lev i et al. noted that the A S S R amplitude increases for sleeping infants for rates between 50-80 H z (Levi et al., 1993). Aoyagi et al. presented A M rates between 20-200 H z to both adults and children and concluded that the optimum rates for detecting the A S S R in children are between 80-100 H z (Aoyagi et al., 1993). A study by Savio et al. tested the hearing of 64 infants between the ages of 0-12 months and found that A S S R at high rates (70-110 Hz) are suitable for threshold testing as early as birth (Savio, Cardenas, Perez Abalo, Gonzales, & Valdes, 2001). Pethe et al. (2004) compared the 40- and 80-Hz A S S R s in children between the ages of 2 months to 14 years old and found that in children under 13 years of age, the 40-Hz A S S R was significantly smaller than the 80-Hz A S S R . Also , the 40-Hz A S S R required a longer period of maturation to reach adult 29 amplitudes, but the 80-Hz A S S R that was more than half the size of the adult response by 1-year of age (Pethe et al., 2004). This makes sense because maturation of generators of the 40-Hz A S S R are delayed, whereas the generators of the 80-Hz A S S R are more adult-like from birth (Levi et al., 1993). To date, there are no studies that examine the effect of maturation on the low-rate A S S R s . Stimulus factors and ASSRs Factors that affect A S S R s include stimulus type, carrier frequency, modulation rate and modulation depth and stimulus intensity. For multiple-ASSRs, the separation between carrier frequencies may also have an effect. Type of stimuli There are several types of stimuli that can elicit A S S R s . These include clicks (Galambos et al., 1981), tone-bursts (Stapells et al., 1984; Stapells et al., 1987), amplitude-modulated tones or noise (Herdman & Stapells, 2001; Lins et al., 1995; Lins & Picton, 1995; Lins et al., 1996), frequency-modulated tones (Picton et al., 1987), and tones that combine A M and F M tones, also called mixed modulation ( M M ) (Dimitrijevic et al., 2001; John et al., 2001; Rickards & Clark, 1984). Carrier frequency The carrier frequencies used to elicit A S S R s typically include 500, 1000, 2000 and 4000 Hz , and sometimes 250, 8000 or 12,000 Hz . The most commonly used frequencies in 30 behavioural audiometric testing. Because one goal of A S S R research is to provide an objective audiometric test, it is important to understand the effect of carrier frequency on this response. Many studies have shown that 80-Hz A S S R s provide an accurate estimation of hearing threshold at the carrier frequencies of 500 through 4000 H z (Dimitrijevic et al., 2001; Dimitrijevic et al., 2002; Herdman & Stapells, 2001, 2003; John et al., 2001; John et al., 1998). The latency and amplitude of the 40-Hz A S S R decrease as carrier frequency is increased (Galambos et al., 1981; Stapells et al., 1984). Similarly, the latency of the 80-Hz A S S R decreases with increasing carrier frequency (Picton et al., 2003). In contrast, the 80-Hz A S S R has larger amplitudes to 1000- and 2000-Hz stimuli, and lower amplitudes to 500- and 4000-Hz stimuli (John et al., 2001). Modulation rate and modulation depth Modulation rate can substantially affect A S S R s . Increasing the modulation rate from low to high changes the amplitude of the response. In general for awake adults, low-rate A S S R s (<20 Hz) have a smaller amplitude that increases up to a rate of 40 Hz , which then decreases significantly (Stapells et al., 1984). The amplitude of the 40-Hz A S S R is increased as the modulation depth is increased from 1% to 100%. However, beyond approximately 50% modulation depth, the amplitude saturates, and the increase in amplitude between 50-100% is not significant (Lins & Picton, 1995; Picton et al., 1987). Modulation rate affects response amplitude differently for sleeping individuals (Cohen et al., 1991; Lev i et al., 1993) and for infants and children (Stapells et al., 1988; Suzuki & Kobayashi, 1984). In these cases, response amplitudes are not necessarily largest for the 40-Hz A S S R ; nevertheless, the 80-Hz A S S R s show an 31 improvement in signal-to-noise ratio (Cohen et al., 1991; Lev i et al., 1993; Stapells et al., 1988; Suzuki & Kobayashi, 1984). Stimulus intensity In general, as the intensity of the stimulus increases, the amplitude of the 40-Hz A S S R increases with increasing intensity up to 90 dB H L , and the latency of the A S S R decreases (Galambos et al., 1981; Regan, 1989; Stapells et al., 1984). A t A M rates around 80 H z , the response also increases with increasing intensity (Lins et al., 1995). However, because the response is smaller than the 40-Hz A S S R , the increase is also proportionally smaller; with a slope of 5-10 nV/dB versus 2 nV/dB for the 40- and 80-Hz responses, respectively (Galambos et al., 1981; Lins et al., 1995; Stapells et al., 1984). Stimulus factors and multiple-ASSRs Carrier frequency The frequencies of interest when using A S S R s as an audiometric tool are typically 500 through 4000 Hz . Continuous, steady-state stimuli, or a successive train of tone-pips, may be used to elicit the A S S R . Brief tonal stimuli may be a better stimulus to elicit the A S S R because it elicits larger A S S R s . However, the response is only significantly larger when brief tonal stimuli are very short (3-4 cycles), and this advantage disappears when the stimulus is presented with other tones (Mo & Stapells, submitted). Also , brief tonal stimuli are not as frequency specific as a sinusoidal A M tone stimulus. For multiple-ASSR, all of these carriers are presented 32 at the same time to one or both ears, effectively testing all frequencies simultaneously (Herdman & Stapells, 2001; John et al., 2003; John et al., 1998; Lins & Picton, 1995). This method is effective as long as there are no significant interactions between the stimuli that increase the time to test and individual (for a detailed description of interactions, see above section titled physiology of interactions). Up to eight A M tones can be presented simultaneously (John et al., 1998). Modulation rate and separation Research on multiple-ASSRs has focussed on the 40-Hz and 80-Hz A S S R s , because both these responses are the most effective measures for threshold testing (Picton et al., 2003). The interactions that occur between responses for multiply presented tones can be affected by modulation rate and the separation between modulation rates of adjacent carrier frequencies. John and colleagues (1998) showed increased interactions for 40-Hz multiple-ASSRs compared to 80-Hz multiple-ASSRs. However, Fontaine and Stapells (in preparation) found that these increased interactions do not preclude the use of the 40-Hz A S S R as an audiologic tool, because multiple stimuli are still more efficient than single stimuli. It is unclear whether interactions can be increased by reducing the separation between A M rates of neighbouring stimuli. John et al. (1998) found that monaural presentation of modulation rates as close as 1.3 H z do not attenuate the mult iple-ASSR in the 80-Hz range, and modulations as close as 4 H z do not attenuate the 40-Hz A S S R in some conditions (John et al., 1998). Van Maanen and Stapells (2005) were able to measure thresholds in individuals using modulation rates as close as 8 H z for the 80-Hz A S S R s and 3 H z for the 40-Hz A S S R s for four stimuli in one 33 ear. Fontaine and Stapells (in preparation) found that response amplitudes were not significantly reduced at 30 dB H L using separations of 3 H z within an ear and separations of 1.5 H z between ears. Modulation separations as high as 7 to 9 H z have been used in studies examining the 80-Hz A S S R thresholds, with no significant effect on ability to obtain responses (Herdman & Stapells, 2001; Lins & Picton, 1995). Overall, the effect of modulation separation has been systematically studied for the 80-Hz A S S R s (John et al., 1998), but not for A S S R s at any other rate ranges. The 40-Hz A S S R s and low-rate A S S R s must be examined for the effects of modulation rate separation in order to determine whether it affects response amplitude of these responses. Stimulus intensity Mul t ip le -ASSR amplitude increases with increasing intensity; however, interactions have been shown to increase as well . Herdman and Stapells (2001) found no significant escalation in interactions when intensity is increased from 30 to 60 dB S P L . However, John et al. (1998) found that interactions for the 80-Hz A S S R s were significantly larger when intensity was increased to 75 dB S P L , decreasing the efficiency of the response. Although amplitude decreased, the mult iple-ASSR remained more efficient than the single stimulus response (John et al., 1998). Fontaine and Stapells (in preparation) demonstrated that at low- to mid-intensities (35 and 55 dB H L ) 8-stimulus 40-Hz multiple-ASSR is more efficient than single A S S R or 4-stimulus multiple-ASSR. However, the authors found that at 80 dB H L , the 8-stimulus 40-Hz mult iple-ASSR was not more efficient than the 4-stimulus multiple-ASSR. 34 Rational for and approach of this study The differences in A S S R s for low-rate (-14 Hz) , mid-rate (40 Hz) and high-rate (80 Hz) A M tones include variation in signal-to-noise levels, differences in where the response originates in the auditory pathways, and differences in interactions between responses in multiply presented stimuli. Both brainstem and cortical generators are active during A S S R stimulation of all rates; however, low- and mid-rate A S S R s have primary generators higher in the auditory pathways, whereas high-rate A S S R s have primary generators in the brainstem (Herdman et al., 2002; Kraus, Ozdamar, Hier, & Stein, 1982; Kuwada et al., 2002; Mauer & Doring, 1999; Palmer, 1995; Rickards & Clark, 1984; Ross et al., 2000; Stapells et al., 1984; Wong & Stapells, 2004). The current A S S R literature is missing information in several areas. First, there is a dearth of information on the low-rate (<20 Hz) A S S R . There is some information on the generators of this response; however it is scant (Herdman et al., 2002). Furthermore, there is no information regarding multiple-ASSRs at this rate and the interactions that may arise as a result. It is necessary that a study be completed that compares the interactions at this rate to those at other rates, as it may offer valuable information on generators at the level of the auditory cortex. Second, there are few studies that investigated the potential usefulness of the 40-Hz multiple-ASSR as a clinical tool. A n d although information is now emerging regarding this technique, more detailed studies of interactions at this rate are needed in order to determine the full extent of the usefulness of the 40-Hz A S S R as a clinical tool and to determine the appropriate clinical protocols. Third, modulation separation between multiple simultaneous stimuli has not been systematically examined for the 40- and 14-Hz A S S R s , and information is needed to determine 35 the effects of separation on response amplitude for A S S R s at all rate ranges. Finally, although much information about interactions of the 80-Hz A S S R exist, there are few studies that directly compare this response and interactions to those at other modulation rates (Van Maanen & Stapells, 2005). Overall, this study w i l l seek to directly compare single and multiple-ASSRs at three levels of the auditory pathway using three different modulation rates: 14, 40, and 80 H z . This w i l l provide information that can be directly compared between rates. This study w i l l also provide much-needed information on the low-rate mult iple-ASSR which, until now, has not yet been studied. 36 Introduction The auditory steady-state response (ASSR) is an auditory evoked potential that has a constant phase and amplitude over time, which is elicited using a repetitive stimulus (Picton et al., 1984). A S S R s are repetitive and they reflect the rate of stimulus presentation. A S S R s can be measured using various carrier frequencies as well as various amplitude modulation rates. Over the years, studies have determined the effects of carrier frequency, as well as amplitude-modulation ( A M ) or frequency-modulation (FM) rate on A S S R s (Aoyagi, Furuse et al., 1994; Campbell et al., 1977; Dimitrijevic et al., 2001; Dobie & Wilson, 1998; John et al., 2003; John et al., 2001; John et al., 1998; Lev i et al., 1993; Lins et al., 1995; Petitot et al., 2005; Picton et al., 1987; Rees et al., 1986; Rickards & Clark, 1984; Ross et al., 2000; Stapells et al., 1984). Of primary interest in the present study are the interactions between stimuli and responses of multiple A S S R s at different levels of the auditory pathways. Mul t ip le-ASSRs are a potentially useful clinical tool; this study wi l l examine the interactions of multiple-ASSRs at three different modulation rates, representing three different levels of the auditory pathways. Neurons in the human auditory pathways wi l l respond differentially depending on the rate of stimulation. More specifically, it has been suggested that low-rate (-14 Hz) A S S R s have at least some generators that are different from mid-rate (<60 Hz) A S S R s , which in turn have different generators than the high-rate 60-Hz A S S R (Herdman et al., 2002; Kuwada et al., 2002; Mauer & Dbring, 1999; Rickards & Clark, 1984; Wong & Stapells, 2004). Generators for the 14-H z A S S R s are primarily in the auditory cortex, 40-Hz A S S R s generators in the auditory cortex and brainstem, and 80-Hz A S S R s generators are primarily in the brainstem. This is a reliable trend that can be observed in A S S R studies of animals (Kuwada et al., 2002; Palmer, 1995; 37 Schreiner & Urbas, 1986; Suzuki, 2002), apparent latencies in human responses (John & Picton, 2000; Kuwada et al., 1986; Rickards & Clark, 1984; Stapells et al., 1984), lesion studies in humans (Spydell et al., 1985; Tachisawa, 1997), and human A S S R magnetoencephalography and electrical source analysis studies (Herdman et al., 2002; Mauer & Doring, 1999; Romani et al., 1982; Ross et al., 2003). A S S R studies at these three rates may provide insight into the different generators in the auditory pathways from brainstem to cortex. The 40-Hz A S S R was the first auditory steady-state response to be established, and was considered a potential tool for audiometric testing in the 1980s (Galambos et al., 1981; Griffiths & Chambers, 1991; Kuwada et al., 1986; Levi et al., 1993; Linden et al., 1985; Mi l ford & Birchall , 1989; Plourde et al., 1991; Stapells et al., 1984; Stapells et al., 1987). The 40-Hz A S S R is a useful audiometric tool because the 40-Hz stimulation produces a larger response than any other modulation rate (Galambos et al., 1981; Ross et al., 2000; Stapells et al., 1984), and it provides a large signal-to-noise ratio (Ross et al., 2000; Stapells et al., 1987). However, the 40-H z A S S R is more difficult to record during sleep, and its amplitude is substantially decreased in infants, an important target group for the audiometric tool (Aoyagi et al., 1993; Stapells et al., 1988; Suzuki & Kobayashi, 1984; Umegaki, 1995). Despite these shortcomings, the 40-Hz A S S R may remain a useful audiologic tool for threshold testing in adults, as a measure of arousal, and as a measure of central auditory integrity (Fontaine & Stapells, in preparation; Jerger et al., 1986; Mi l ford & Birchall , 1989; Petitot et al., 2005; Plourde et al., 1991; Van Maanen & Stapells, 2005). The 40-Hz A S S R is affected by sedation or state of arousal, as well as by the age of the subject, but this is not the case for A S S R s at higher modulation rates (80-110 Hz) (Aoyagi, 38 Furuse et al., 1994; Aoyagi et al., 1993; Cohen et al., 1991; John et al., 2003; John et al., 1998; Lins & Picton, 1995). In recent years, the 80-110 H z A S S R ("80-Hz A S S R " ) has been shown to be a useful measure of audiometric thresholds in normal hearing adults (Aoyagi, Kiren et al., 1994; Dimitrijevic et al., 2002; Herdman & Stapells, 2001; Lins et al., 1996; Perez-Abalo et al., 2001; Ranee et al., 1995), infants and children (Aoyagi, Kiren et al., 1994; Lins et al., 1996; Perez-Abalo et al., 2001; Ranee & Rickards, 2002; Ranee et al., 1995), high-risk newborns (Luts et al., 2006), and those with hearing impairment (Aoyagi, Kiren et al., 1994; Dimitrijevic et al., 2002; Han et al., 2006; Herdman & Stapells, 2003; Lins et al., 1996). Therefore, it seems that the 40-Hz and 80-Hz A S S R s may both have important future roles clinical testing for adults and children. L o w rate (-14 Hz) A S S R s have not been extensively studied as a potential audiometric tool because this response can be difficult to obtain, and it may be difficult to separate from the background E E G noise (Herdman et al., 2002; Picton et al., 1987). E E G noise levels are higher at lower frequencies, such as 14 Hz , compared to those around 40 or 80 H z (Herdman et al., 2002; Picton et al., 2001). However, it is not impossible to record responses at low rates, and these responses have been shown to provide insight into higher auditory processes that cannot be provided by higher modulation rates (Dajani & Picton, 2005; Stefanatos, 1993; Stefanatos et al., 1997; Stefanatos et al., 1989; Wong & Stapells, 2004). For example, one study found that a variable-length 1000-Hz carrier tone that was frequency modulated at 4 H z could be used to detect dysfunction in auditory mechanisms responsible for perception of a F M signal in children (Stefanatos et al., 1997). A n earlier study found that low-rate A S S R s to F M tones may be able to identify dysfunction of auditory receptive abilities in children with Landau-Kleffner syndrome 39 (Stefanatos, 1993). In another study, Wong and Stapells (2004) compared the brainstem and cortical A S S R binaural masking level difference ( B M L D ) using 13- and 80-Hz A M tones. The B M L D existed for the low-rate A S S R but not the high-rate A S S R , indicating a difference in binaural processing (and thus, generators) between the two A S S R s (Wong & Stapells, 2004). Taken together, these studies indicate that the low-rate A S S R may provide information about the auditory pathways that is clinically relevant, such as for higher auditory processing disorders. However, there is a dearth of information on the low-rate A S S R alone or in comparison to other rates. A S S R s are a useful audiological technique because they can be recorded simultaneously to multiple stimuli, therefore decreasing the time required to test an individual (Lins et al., 1995; Lins & Picton, 1995). First introduced by Regan in vision studies, responses can be recorded to more than one stimulus simultaneously, provided that each stimulus is modulated at a different rate (as reviewed by Regan, 1989). Most mult iple-ASSR studies have used the 80-Hz A S S R due to its ability to obtain responses in infants and sleeping adults (Cone-Wesson et al., 2002; Dimitrijevic et al., 2004; Dimitrijevic et al., 2001; John et al., 2001; John et al., 1998; Johnson & Brown, 2005; Lins et al., 1995; Lins & Picton, 1995). However, new research is now emerging that the A S S R to multiple stimuli may be useful in the 40-Hz range (Fontaine & Stapells, in preparation; Van Maanen & Stapells, 2005) and, in fact, may be more efficient at measuring thresholds in adults (Fontaine & Stapells, in preparation; Petitot et al., 2005; Van Maanen & Stapells, 2005). Few studies have directly compared 40- and 80-Hz A S S R (Petitot et al., 2005; Van Maanen & Stapells, 2005), and no study has examined the use of multiple stimuli on the low-rate A S S R . Most importantly, no studies have compared A S S R s at all three modulation 40 rates. Each response may provide useful clinical information, but direct comparison among them wi l l provide additional information about neural mechanisms, or highlight the specific clinical uses of A S S R s at each rate. When presenting multiple stimuli simultaneously, interactions between responses to the stimuli occur in the A S S R recordings. A response interaction is defined as a change (usually a decrease) in response amplitude as a result of the presentation of more than one tone simultaneously in either one or both ears. The 80- and 40-Hz A S S R interactions have different response amplitude decreases, and different contributing mechanisms (Fontaine & Stapells, in preparation; John et al., 1998). In addition to cochlear interactions, some of these differences may be a result of mechanisms at the levels of brainstem and/or the cortex (John et al., 1998). This hypothesis may be clarified by including research on interactions for multiple-ASSRs at low rates. A more thorough picture of all interactions would result from the systematic manipulation of the parameters that influence these responses both between multiple-ASSRs at 40- and 80-Hz and regarding interactions for multiple A S S R s at low modulation rates. The known differences between 40- and 80-Hz A S S R s suggest that variation in interactions may be a result of these responses arising from different levels of the auditory pathways. Some of the generators of low-rate A S S R s arise from the cortex (Herdman et al., 2002); therefore, the present study wi l l provide new insight into higher level interactions that have not yet been determined. Variation in such interactions may indicate differences in processing along the auditory pathways at each rate range, and may provide insight into the clinical usefulness of each modulation rate. 41 Methods Subjects Twelve adult subjects participated in this study (7 female; mean age for all subjects: 21.1 years). A l l subjects were screened behaviourally for normal hearing (<20 dB H L ) for pure tones at 500, 1000, 2000, and 4000 Hz . Subjects provided informed consent and were paid an honorarium for their participation. A n additional 6 subjects were rejected as follows: one subject was rejected due to a long-standing hearing loss that was not disclosed at the beginning of the session. This subject was advised to see an audiologist for further evaluation. Five other subjects were rejected due to responses that were below a previously specified amplitude criterion in the 14-Hz single-tone monotic condition (responses had to be ^90 nV) . This was expected, as pilot studies suggested approximately 25% of subjects may have very small 14-Hz A S S R s , who would be rejected because of excessively long testing time required to obtain significant responses. Stimuli Stimuli were sinusoidal 100% amplitude-modulated tones with carrier frequencies of 500, 1000, 2000 and 4000 Hz . There were three main sections of the study, divided according to modulation rate: 14, 40 and 80 Hz . The 1000-Hz carrier served as a baseline condition for the experiment. Amplitude modulation of the 1000-Hz tone remained constant at all three rates at either: 13.672, 40.039 or 80.078 H z (nominally, 14, 40 and 80 Hz). Each modulation rate had four conditions, including a single 1000-Hz tone ( M l , baseline), a two-tone dichotic condition (D2), a two-tone monotic condition (M2), and a four-tone monotic condition (M4). Amplitude 42 modulations of the 500, 2000 and 4000 H z tones were varied depending on the separation of the A M rates between carriers for multiple-stimuli conditions (either 2, 4 or 6 Hz). There were 30 conditions in total, 10 conditions for each of 14-, 40- and 80-Hz stimuli. Results were only measured in response to the 1000-Hz stimuli. The left ear was used as the test ear for all subjects. Tables 1-3 provide a full description of modulation rates at each carrier frequency. The order of the conditions within each section was randomized; and the order of the 14-40- and 80-Hz sections was also randomized. However, there were several instances where a modulation rate was changed before all the conditions were completed within a section. This occurred only when the subject began to fall asleep, and 14- and 40-Hz conditions would be affected. For these participants, stimuli were changed to the 80-Hz modulation rate and incomplete conditions were presented later when the subject was awake. Table 1. Modulation Frequencies of 14-Hz Stimuli C O N D I T I O N S E P A R A T I O N (Hz) 500 H z 1000 H z 2000 H z 4000 H z Monotic Single ( M l ) N / A 13.672 2-Tone Dichotic (D2) 2 13.672 15.625 4 13.672 17.578 6 13.672 19.531 2-Tone Monotic (M2) 2, 13.672 15.625 4 13.672 17.578 6 13.672 19.531 4-Tone Monotic (M4) 2 11.719 13.672 15.625 17.578 4 9.766 13.672 17.578 22.461 6 7.812 13.672 19.531 26.376 43 Table 2. Modulation Frequencies of 40-Hz Stimuli C O N D I T I O N S E P A R A T I O N (Hz) 500 H z 1000 H z 2000 H z 4000 H z Monotic Single ( M l ) N / A 40.039 2-Tone Dichotic (D2) . 2 40.039 41.992 4 40.039 43.945 6 40.039 45.898 2-Tone Monotic (M2) 2 40.039 41.992 4 . 40.039 43.945 6 40.039 45.898 4-Tone Monotic (M4) . 2 38.086 40.039 41.992 43.945 4 36.133 40.039 43.945 47.842 6 34.18 40.039 45.898 51.758 Table 3. Modulation Frequencies of 80-Hz Stimuli C O N D I T I O N S E P A R A T I O N (Hz) 500 H z 1000 H z 2000 H z 4000 H z Monotic Single ( M l ) N / A 80.078 2-Tone Dichotic (D2) 2 80.078 82.031 4 80.078 83.984 6 80.078 85.937 2-Tone Monotic (M2) 2 80.078 82.031 4 80.078 83.984 6 80.078 85.937 4-Tone Monotic (M4) 2 78.125 80.078 82.031 83.984 4 76.172 80.078 83.984 87.891 6 74.219 80.078 85.937 91.797 44 The A M stimuli were calibrated in dB S P L , with the level of each individual stimulus presented at 80 dB SPL . Stimuli were presented via air-conduction through E R - 3 A insert earphones. Modulation rates and carrier frequencies were chosen so that each recording sweep of the E E G would contain an exact integer number of cycles for both carrier and A M rate. A l l tones were created by the Rotman Multiple Auditory Steady-State Evoked Response ( M A S T E R ) program (John et al., 1998), and were attenuated through Tucker-Davis Technologies (TDT) P A 1 attenuators, then routed to a T D T H B 6 module. Acoustic calibration of the signal was checked throughout the experiment using a Quest Model sound level meter and a Bruel and Kjaer DBO138 2cc coupler. Electroencephalogram (EEG) recording Gold-plated recording electrodes were placed on a subject's forehead (ground), ipsilateral (left) earlobe (inverting), and vertex (non-inverting). Inter-electrode impedance was kept at or below 5 kOhms at 10 H z for the recording. For the 14-Hz A S S R S , the E E G was bandpass filtered between 1-30 H z (12 dB/octave), and amplified with a gain of 40,000. For the 40- and 80-Hz A S S R s , the E E G was bandpass filtered between 5-100 (12 dB/octave) and 30-250 (12 dB/octave), respectively. Total amplifier gain for the 40- and 80-Hz conditions was 80,000. The E E G was recorded and averaged using a sweep that consisted of 16 epochs. Each epoch contained windows that were 1.024 seconds long; therefore, each sweep was 16.38 s. The number of sweeps differed for each modulation rate, but could be as few as 10 sweeps, or as many as 40. 45 ASSR analysis An EEG recording sweep of 16.38 s (sampling rate of 1000 Hz) was used for all ASSRs. The ASSRs were analysed in the frequency domain by FFT. The FFT resolution was 0.083 Hz, spanning from 0 to 250 Hz. Peak-to-peak amplitudes were measured and reported as amplitude. Determination of significance was made by on-line analysis of variance (F-statistic) (Lins et al., 1996). The signal-to-noise ratio (F-ratio) was calculated by testing the amplitude of the modulation frequency of interest against the amplitude of the 120 surrounding frequency bins (60 above and 60 below the modulation frequency). Thus, the frequency bins extend 4.98 Hz on each side of the modulation frequency. A "response" was determined to be significant when the F-ratio of the signal-to-noise was at p<.05. A "no response" was determined if the EEG background noise met both the noise criterion, and p>.05. The noise criterion was determined by the circle radius (CR), which represents the 95% confidence limits for the averaged noise (John & Picton, 2000). In recording with no response, recording was continued until the CR was less than 20 nV for the 80-Hz ASSR (Dimitrijevic et al., 2002; Herdman & Stapells, 2003; Small & Stapells, 2005a), less than 60 nV for the 40-Hz ASSR (Van Maanen & Stapells, 2005), and less than 150 nV for the 14 Hz ASSR. The noise criterion for the 14-Hz ASSR was determined based on pilot studies, as well as data previously obtained by Wong and Stapells (2004). In addition to significance and noise stopping rules, a minimum of 10 sweeps and a maximum of 40 sweeps were used. To obtain a measure of relative efficiency, allowing statistical comparisons of different conditions, the relative efficiency of each subject's amplitudes were calculated. Individual response amplitudes for a particular modulation rate and condition were divided by the mean of 46 all subjects' responses to the corresponding modulation rate and condition. This normalized the multiple-response conditions (D2, M 2 , M 4 ) to the baseline condition ( M l ) . The normalized M 4 conditions were multiplied by two (v/4, representing four times the information recorded at once compared to the M l condition) and the D2 and M 2 conditions were multiplied by 1.41 (>/2, representing two times more information being recorded compared to the M l condition). These values were averaged across subjects. Data analysis and statistical analysis Amplitudes and relative efficiencies of the 14-, 40- and 80-Hz responses were analyzed. First, a 3-way repeated-measures analysis of variance ( A N O V A ) determined the effect of separation of modulation rates for all modulation rates for all conditions (except the baseline), with amplitude modulation rate (3 levels), number of stimuli (3 levels), and separation of tones within each rate range (3 levels) as within-subject factors. Amplitudes were subsequently pooled over separation and a 2-way repeated-measures A N O V A completed to determine the effect of condition (4 levels) and modulation rate range (3 levels), as well as the interaction between these two factors. Finally, three one-way repeated-measures A N O V A s were completed for 14-, 40-and 80-Hz relative efficiency data with condition (4 levels) as a factor, to determine the effect of number of stimuli on relative efficiency for each modulation rate range separately. Huynh-Feldt epsilon corrections were used as appropriate for all repeated-measures A N O V A s . Neuman-Keuls post-hoc analysis were completed for significant main effects and interactions. Results were considered significant at p<.05 for all tests. 47 Procedure The study involved one recording session totalling 2.5-4.5 hours or two sessions of 1.5-2 hours each. Eleven subjects completed the study in one session, one subject completed the study in two sessions. Before the first session, participants were screened for normal hearing. If subjects returned for a second session, screening was completed again to ensure no change in thresholds between sessions. A l l subjects sat in a reclining chair in a sound-attenuated booth throughout the recordings. They were invited to sleep during the 80-Hz recordings and otherwise watched a silent closed-captioned movie of their choice for 14- and 40-Hz recordings. A l l subjects were able to relax for most of the testing and some of the subjects slept during the 80-Hz recordings. This research was approved by the University of British Columbia Behavioural Research Ethics Board (see Appendix E) . 48 Results Responses were obtained for the test frequency (1000 Hz) at all modulation rates and for all conditions. Figure 3 shows Fast Fourier Transforms (FFT) of responses of a typical subject to 4 conditions ( M l , M 2 , D2 , M 4 ) at each modulation rate (14, 40 and 80 Hz) , for one modulation rate separation (2 Hz) . This figure indicates that the 40-Hz A S S R was had a larger amplitude than both the 14- and 80-Hz A S S R s . Also , the figure illustrates the general trend of smaller response amplitudes for multiple stimuli compared to single stimuli. Furthermore, there are different amplitude decrease patterns for each modulation rate. Mean A S S R amplitudes for all subjects grouped by modulation rate for all conditions are presented in Table 4. There is no clear effect of modulation rate separation (i.e., 2 vs 4 vs 6 Hz) , as confirmed by a 3-way repeated-measures A N O V A of all amplitude data with modulation rate (3 levels), condition (4 levels) and separation (3 levels) as factors (the 1-tone monotic ( M l ) condition was excluded). N o significant main effects or interactions involving rate separation were found (p-values .248-.311). Because separation had no effect on responses, amplitudes were pooled over all separations (2,4 and 6 Hz) , resulting in one amplitude for each condition at each rate for each subject. Pooled amplitudes for all conditions at all rates are shown in Figure 4. The figure is grouped according to modulation rate (14, 40, 80 Hz) , and within modulation rate the results are grouped by condition ( M l , D2 , M 2 , M4) . Figure 4 demonstrates that the amplitude of the 1000-Hz responses varied depending on the condition as well as the modulation rate. 49 CONDITION 1-Tone Monotic 2-Tone Dichotic 40-Hz ASSR 80-Hz ASSR 100 nVl 2-Tone Monotic 50 nV] \ \ T \ • • 4-Tone Monotic 10 14 18 22 26 30 20 30 40 50 60 70 80 90 100 110 EEG FREQUENCY (Hz) Figure 3. Fast Fourier Transforms (FFT) of ASSRs at all conditions at all rates at a separation of 2 H z for a typical subject. Shown are amplitude spectra resulting from F F T analyses (10-30 Hz for 14-Hz A S S R s , 20-60 Hz for 40-Hz A S S R s and 70-110 Hz for 80-Hz ASSRs) . The test carrier frequencies (1000 Hz) for all modulation rates are represented by arrows and non-test carrier frequencies (500, 2000 and 4000 Hz) are represented by triangles. Fi l led symbols represent a response that is significantly different from the background noise (p<.05), and open symbols represent non-significant responses. Table 4. Mean and Standard Deviation Amplitudes (nV) to 1000-Hz Stimuli PRESENTATION SEPARATION 14 Hz 40 Hz 80 Hz T Y P E Monotic Single ( M l ) .' N / A . 192 ± 6 2 326 ± 1 2 9 180 ± 7 1 2-Tone Dichotic (D2) 2 H z 155 ± 5 7 246 ± 101 158 ± 7 8 4 H z 182 ± 9 4 253 ± 1 0 5 147 ± 7 7 6 H z 184 ± 8 7 269 ± 1 0 5 156 ± 8 5 2-Tone Monotic (M2) 2 H z 171 ± 6 0 205 ± 8 1 98 ± 4 5 4 H z 196 ± 8 6 201 ± 86 87 ± 4 1 6 H z 173 ± 7 2 212 ± 8 3 100 ± 54 4-Tone Monotic (M4) 2 H z 103 ± 5 4 127 ± 5 1 101 ± 3 2 4 H z 87 ± 3 9 128 ± 67 92 ± 3 5 6 H z 117 ± 9 5 130 ± 64 86 ± 2 9 Amplitudes of the 1000-Hz response decreased as the number of stimuli increased; however, the decrease was not consistent across modulation rates or between conditions within a single modulation rate. A two-way repeated-measures A N O V A revealed main effects for modulation rate (3 levels) and condition (4 levels) as factors and.an interaction between the two factors (Table 5). The amplitude of the 40-Hz A S S R s , pooled across condition, were significantly larger than the 14- and 80-Hz responses (p<.003 and p<.001, respectively). There were no significant differences between the amplitudes of the 14- and 80-Hz responses (p=.098). 51 CL < 500 400 > UJ 300 Q 200 100 I I Monotic single (M1) I I Dichotic two (D2) • • • Monotic two (M2) Monotic four (M4) 14 40 MODULATION RATE (Hz) 80 Figure 4. Mean ASSR amplitudes for all conditions at 14, 40 and 80 Hz, pooled over separation of modulation rates; error bars represent 1 standard deviation. Table 5. Summary Source Table of two-way repeated-measures ANOVA for amplitude measurements comparing effects of modulation rate and condition EFFECT df Modulation Rate 2, 22 9.3 0.844 p<.001 Condition 3, 33 61.28 1 p<.001 Modulation Rate X Condition 6, 66 13.56 0.868 p<.001 df = degree of freedom aHyunh-Feldt epsilon (e) correction factor for degrees of freedom bProbability reflects corrected degrees of freedom * significant (p<.05) 52 In the 80-Hz A S S R condition, amplitudes were reduced from baseline (1-tone condition, M l ) only for the 2-tone monotic (M2) and 4-tone monotic (M4) conditions (p<.001 and .001, respectively); the 2-tone dichotic condition (D2) did not differ from baseline (p=.460). Both the M 2 and the M 4 conditions had amplitudes that were smaller than the D 2 condition (p=.003 for both) for the 80-Hz condition; they were not significantly different from each other (p=.911). The 40-Hz multiple-ASSR amplitudes were significantly reduced from the M l condition for all multiple tone conditions, D2 , M 2 and M 4 (p<.001 for all). Amplitudes were reduced from the D2 condition when compared to the M 2 (p<.001) and M 4 (p<.001) conditions. Finally, M 4 was significantly smaller than the M 2 condition (p=.011). Unlike the 40- and 80-Hz multiple A S S R s , the 14-Hz A S S R amplitudes were reduced only in the M 4 condition (p<.001). The M 4 condition was also significantly smaller than the D 2 and M 2 conditions (p<.001 for both). Neither of the remaining two conditions (D2 and M 2 ) were reduced compared with the baseline (p =.81 and .75, respectively); and they were not different from each other (p =.927). Relative efficiency results, shown in Figure 5, were.analysed using one-way repeated-measures A N O V A s for each modulation rate separately, with condition (4 levels) as a factor; thus allowing a direct comparison among conditions within a modulation rate. Relative efficiencies were not compared across modulation rates. Newman-Keuls post-hoc analysis of relative efficiencies indicated that none of the multiple-ASSRs were more efficient than the baseline except monotic 2-tone condition for the 14-Hz A S S R (p=.032; 1.00 vs. 1.32). For the 14-Hz A S S R , the D2 and M 4 conditions were not significantly different from baseline (p=.072 and .738, respectively). For the 40-Hz A S S R , none of the conditions were different from baseline (D2: 53 p=.805, M 2 : p=.563, M 4 : p=.181). Finally, for the 80-Hz A S S R , none of the conditions are significantly different from baseline (D2 p=.428, M 2 p=.072, M 4 p=.986). o z UJ o 2.0 UJ 0.5 Monotic single (M1) Dichotic two (D2) Monotic two (M2) Pv^ l Monotic four (M4) 14 40 80 MODULATION RATE (Hz) Figure 5. Mean relative efficiencies for all conditions at 14, 40 and 80 H z ; error bars represent 1 standard deviation. 54 Discussion Modulation rate, number of stimuli, and monotic versus dichotic stimulation influence interactions, and consequently the amplitudes, of multiple-ASSRs. However, there was no significant effect of modulation rate separation between stimuli for any of the modulation rates studied. A s expected, the 40-Hz single A S S R was significantly larger than the 14- and 80-Hz A S S R s , a finding that is consistent with previous findings (Galambos et al., 1981; Ross et al., 2000; Stapells et al., 1984). Within each modulation rate, the different conditions displayed a general trend for amplitudes to decrease from a single tone ( M l ) to multiple tones (D2, M 2 , M4) . This has also been reported to some degree in the literature (Fontaine & Stapells, in preparation; John et al., 1998). However, each rate range demonstrated a different pattern of amplitude decrease over these conditions. In general, the 40-Hz A S S R s had amplitudes that were reduced in the D2 , M 2 and M 4 conditions, with no significant improvement in relative efficiency as number of stimuli increased. The 80-Hz A S S R s had amplitudes reduced in the M 2 and M 4 conditions, when compared to the baseline, also with no improvement in relative efficiency with increasing number of stimuli. The 14-Hz A S S R departs from this trend, with a decrease in amplitude only in the M 4 condition and a significant improvement in relative efficiency for the M 2 condition only. John and colleagues noted different results from the present study for the 40-and 80-Hz A S S R s . When using a 1000-Hz baseline condition, John et al. (1998) found that two tones did not cause any significant interactions for the 40-Hz A S S R , and four tones did not cause any significant interactions for the 80-Hz A S S R (at 60 dB SPL) . When the intensity level was increased to 75 dB S P L , the authors found that interactions did exist for the 80-Hz A S S R . Results from their study suggest that 80-Hz multiple A S S R s at 75 dB S P L are neither more nor 55 less efficient than a single-stimulus presentation, although this was not tested statistically. The current study presented stimuli at 80 dB S P L and found interactions at all modulation rates, which is consistent with the overall pattern of intensity results found by John and colleagues (1998). Other studies have shown that A S S R interactions are smaller at lower intensities (Fontaine and Stapells, in preparation; Herdman and Stapells, 2001). Herdman arid Stapells (2001) found that increasing intensity from 30-60 dB S P L had little effect on the A S S R interactions, and Fontaine and Stapells (in preparation) demonstrated that interactions are not as large at 30 and 55 dB H L compared to 80 dB H L . More specifically, the pattern of 40-Hz A S S R interactions found by Fontaine and Stapells (in preparation) are similar to those found in the current. In general, the previous literature shows that increasing intensity results in increased interactions (Fontaine and Stapells, in preparation; John et al., 1998). The current results are consistent with these findings. This suggests that multiple-ASSRs to high-intensity stimuli are no more efficient than for a single stimulus, for all modulation rates. The present study demonstrates that mult iple-ASSR interactions are different at each modulation rate, with different patterns of interactions for each modulation rate, which can be taken as further evidence for different underlying neural generators for the 14-, 40-, and 80-Hz A S S R s (Picton & Stuss, 1980). It appears that the three rate "regions" (low-, mid- and high-frequency ASSRs) reflect processes at different levels in the auditory pathways (brainstem, primary auditory cortex, probably other cortical areas, with generators different for 14 and 40 Hz). That is not to say that each rate region is processed independently of the others; simply that at least some of the generators at each rate range differ. In general, the largest interactions occur in the 40-Hz A S S R s , and the smallest interactions occur in the 14-Hz A S S R s . Although there is 56 a difference in interactions between rate regions, it not does seem to affect the relative efficiencies. In other words, at 80 dB S P L , for each of the three modulation rates, presentation of multiple stimuli did not significantly improve the efficiency of testing. Nevertheless, the results show that interactions increase as we move from the brainstem to the primary auditory cortex, but then decrease as other cortical auditory areas are included. Some of these interactions occur in the cochlea and some do not. For monotic stimuli, interactions could be occurring in the cochlea, beyond the cochlea, or both. However, for dichotic stimuli, interactions cannot be occurring in the cochlea and must be occurring beyond the level of the 8 t h cranial nerve and cochlear nuclei. John and colleagues hypothesized that there are factors affecting the interactions between multiple stimuli, including: 1) basilar membrane travelling wave interactions, 2) compressive rectification, and 3) two-tone suppression (for a description of these factors, see John et al., 1998). The differences found between modulation rate regions in the current study and in John et al. (1998) demonstrate that the above three factors cannot account for all interactions. John and colleagues describe the remaining interactions as being a result of "residual" effects. The present study indicates that there may be more than one "residual" effect. For example, the 14-Hz A S S R s have significant interactions only in the four-tone condition, whereas the 40-Hz A S S R s have significant interactions for all multiple-tone conditions. Both of these responses are occurring in the cortex, but they have different patterns of interactions. Also , a significant interaction of the dichotic 2-tone condition occurs only for the 40-Hz A S S R , but not in the 14- or 80-Hz A S S R . There are, therefore, various "residual" effects, and they differ for each modulation rate region. 57 Clinical Implications Overall, it appears that interactions in the present study are larger than in previous studies for both the 40- and 80-Hz A S S R s . The primary reason is that, unlike previous studies, the current study employed a relatively high intensity (80 dB SPL) . This is relevant clinically, because the multiple-ASSR is used to assess hearing losses in both children and adults. If large interactions occur at high intensities, then the multiple-ASSR may not be a more efficient test for those with large hearing losses. Further studies are necessary to determine the effects of severe to profound losses on the accuracy and efficiency of the multiple-ASSR. Similar to the 80-Hz A S S R , the 40-Hz multiple A S S R may be useful as a practical and objective measure of audiometric thresholds. Because this response is attenuated in children and sleeping adults (Aoyagi et al., 1993; Stapells et al., 1988; Suzuki & Kobayashi, 1984; Umegaki, 1995), this response would not be practical for use for these groups. However, the 40-Hz A S S R may be an ideal objective threshold measure for awake adults, because this response has larger amplitudes compared to other modulation rates for awake adults, and has been shown to be a faster and more accurate depiction of threshold than the 80-Hz mult iple-ASSR (Petitot et al., 2005; Van Maanen & Stapells, 2005). There have been few studies for the 40-Hz mult iple-ASSR relative to the 80-Hz multiple-ASSR, and more studies must be completed before this measure can be used as a standard tool in the audiometric test battery. It is possible that the 40-Hz A S S R may be useful in determining location of damage or lesions in the auditory pathway. Studies have shown that the 40-Hz A S S R is reduced in patients that have brainstem or thalamic lesions (Firsching et al., 1987; Harada et al., 1994; Spydell et al., 1985). However, the same amplitude reduction in the response does not exist in patients who 58 have temporal lobe lesions unilaterally (Firsching et al., 1987; Spydell et al., 1985). Also , an absent 40-Hz A S S R has been correlated with brain death, therefore, it may be useful in determining prognosis or recovery of brain injured comatose patients (Firsching et al., 1987). Another clinical application for the 40-Hz A S S R is its usefulness in measuring arousal levels of patients under general anaesthesia. Plourde et al. (1991) found that sleep attenuated the 40-Hz A S S R and sedation attenuated it even further. This and other studies suggest that the 40-H z A S S R makes a safe and accurate measure of patient arousal during sedation since the response attenuates with sedation and subsequently increases with awakening (Plourde & Picton, 1990; Plourde et al., 1991; Plourde & Villemure, 1996). There are several interesting clinical implications for the low-rate A S S R s , because they likely represent processing at higher levels of the auditory pathways including the cortex and potentially auditory association areas. Studies by Stefanatos and others have correlated low-rate A S S R s with auditory processing deficits in children. Wong and Stapells (2004) found a difference in the binaural masking level difference ( B M L D ) of 13- and 80-Hz A S S R s . The B M L D does not appear to exist in the 80-Hz A S S R , but it does in the 14-Hz A S S R . This is similar to transient response results obtained from other studies on the B M L D , and suggests binaural processing is different at the level of the brainstem compared to the cortical level. Using multiple-ASSRs at different modulation rates together may also have clinical implications. Because modulation rates used together provide different patterns of interactions, these patterns of interactions could be used in patients to potentially determine central auditory processing deficits. 59 Conclusion There is now a body of evidence demonstrating that A S S R s are a reliable, effective, and objective measure of thresholds in normal and hearing-impaired individuals, as well as children, infants and those who are sleeping or sedated (Aoyagi, Furuse et al., 1994; Dimitrijevic et al., 2002; Griffiths & Chambers, 1991; Herdman & Stapells, 2001,2003; Lins et al., 1996; Mi l ford & Birchall , 1989; Perez-Abalo et al., 2001; Picton et al., 1998; Ranee & Rickards, 2002; Ranee et al., 1995; Swanepoel et al., 2004; Van Maanen & Stapells, 2005). This method of threshold estimation is time efficient, accurate, and does not require the subjective judgements of a clinician. It has also been shown that A S S R s can be used to determine aided thresholds in individuals with hearing aids or cochlear implants (Picton et al., 1998). A S S R s can also be used as a measure to determine central auditory integrity such as identifying dysfunction of auditory receptive abilities in children (Stefanatos, 1993; Stefanatos et al., 1997), potentially obtaining hearing thresholds in those with auditory neuropathy (Picton et al., 1998; Ranee et al., 1999), predicting a person's word recognition scores (Dimitrijevic et al., 2004), predicting the ability to listen in noise using the binaural masking level difference ( B M L D ) for low-rate A S S R s (Wong & Stapells, 2004), or measuring of integrity of the auditory pathway after brain trauma (Tachisawa, 1997). Furthermore, the 40-Hz A S S R is affected by both sleep and anaesthesia, and it has been suggested that this response would be a good measure of level of sedation (Plourde et al., 1991), and the low-rate A S S R B M L D is also thought to provide an early indication of auditory brainstem dysfunction or auditory processing disorders, that cannot be identified using hearing thresholds (Musiek & Lamb, 1994; Wong & Stapells, 2004). In general, the 80-Hz responses would indicate functioning up to the level of the brainstem, and 60 the 40-Hz or low-rate responses would indicate functioning up to the level of the auditory cortex or auditory association areas. Studying A S S R interactions provides information on the functioning of the auditory pathways; and therefore, may be useful in determining the different processing between those with normal hearing and those who have cortical re-organization due to hearing loss. The present study demonstrates that separation between modulation rates in multiple-A S S R s does not affect the response amplitudes. However, it also demonstrates that adding multiple stimuli, especially four tones, increases interactions enough that multiple tone presentation may no longer be more efficient than presenting a single tone. This study presented stimuli at a relatively high level (80 dB SPL) , so greater interactions were expected. However, -this calls into question the use of mult iple-ASSR as the most efficient method of objective audiometry when stimuli are presented at a high level (e.g., for those with severe or profound hearing loss). In conclusion, there are many potential uses for A S S R s , although more research is needed before multiple-ASSRs can replace current methodologies as the standard objective electrophysiological measurement of hearing. The current research adds to the wealth of information that already exists, and provides a step in the right direction for establishing A S S R s as the standard objective measure of audiometry for both children and adults. 61 Bibliography Aoyagi, M . , Furuse, H . , Yokota, M . , Kiren, T., Suzuki, Y . , & Koike , Y . (1994). Detectability of amplitude-modulation following response at different carrier frequencies. Acta Otolaryngologica Supplement, 511, 23-27. Aoyagi, M . , Kiren, T., Furuse, H . , Fuse, T., Suzuki, Y . , Yokota, M . , & Koike , Y . (1994). Pure-tone threshold prediction by 80-Hz amplitude-modulation following response. Acta Otolaryngologica Supplement, 511,1-14. Aoyagi, M . , Kiren, T., K i m , Y . , Suzuki, Y . , Fuse, T., & Koike, Y . (1993). Optimal modulation frequency for ampltude-modulation following response in young children during sleep. Hearing Research, 65,253-261. Campbell, F . W. , Atkinson, J., Francis, M . R., & Green, D . M . (1977). Estimation of auditory thresholds using evoked potentials. A clinical screening tool. Progress in Clinical Neurophysiology, 2, 68-78. Cohen, L . T., Rickards, F. W . , & Clark, G . M . (1991). A comparison of steady-state evoked potentials to modulated tones in awake and sleeping humans. Journal of the Acoustical Society of America, 90(5), 2467-2476. Cohen, M . M . (1982). Coronal topography of the middle latency auditory evoked potentials ( M L A E P s ) in man. Electroencephalography and Clinical Neurophysiology, 53, 231-236. Cone-Wesson, B . , Dowel l , R. C , Tomlin, D . , Ranee, G . , & M i n g , W . J. (2002). The auditory steady-state response: Comparisons with the auditory brainstem response. Journal of the American Academy of Audiology, 13, 173-187. 62 Dajani, H . R., & Picton, T. W . (2005). Auditory steady-state responses to noise stimuli with changing inter aural correlation. Paper presented at the International Evoked Response Audiometry Study Group, Havana, Cuba. (June 12-16, 2005). Dimitrijevic, A . , John, M . S., & Picton, T. W . (2004). Auditory steady-state responses and word recognition scores in normal-hearing and hearing-impaired adults. Ear and Hearing, 25, 68-84. Dimitrijevic, A . , John, M . S., van Roon, P., & Picton, T. W . (2001). Human auditory steady-state responses to tones independently modulated in both frequency and amplitude. Ear and Hearing, 22, 100-111. Dimitrijevic, A . , John, M . S., van Roon, P., Purcell, D . W. , Adamonis, J. , Ostroff, J., Nedzelski, J. M . , & Picton, T. W . (2002). Estimating the audiogram using multiple auditory steady-state responses. Journal of the American Academy of Audiology, 13, 205-224. Dobie, R. A . , & Wilson, M . J. (1998). Low-level steady-state evoked potentials: Effects of rate and sedation on detectability. Journal of the Acoustical Society of America, 104(6), 3482-3488. Dolphin, W . F i ; & Mountain, D . C . (1993). The envelope following response (EFR) in the Mongolian gerbil to sinuasoidally amplitude-modulated signals in the presence of simultaneously gated pure tones. Journal of the Acoustical Society of America, 94, 3215-3226. Don, M . , Elberling, C , & Waring, M . (1984). Objective detection of averaged auditory brainstem responses. Scandinavian Audiology, 13, 219-228. 63 Elberling, C , & Don, M . (1984). Quality estimation of averaged auditory brainstem responses. Scandinavian Audiology, 13, 187-197. Firsching, R., Luther, J., Eidelberg, E . , Brown, W . E . , Jr., Story, J. L . , & Boop, F . A . (1987). 40 Hz-middle latency auditory evoked response in comatose patients. Electroencephalography and clinical Neurophysiology, 67, 213-216. Fontaine, C , & Stapells, D . R. (in preparation). Efficiency of single versus multiple stimuli for 40-Hz auditory steady-state responses. Unpublished M . S c . Thesis, University of British Columbia. Galambos, R., Makeig, S., & Talmachoff, P. J. (1981). A 40-Hz auditory potential recorded from the human scalp. Proceedings of the National Academy of Science (USA), 78, 2643-2647. Griffiths, S. K . , & Chambers, R. D . (1991). The amplitude modulation-following response as an audiometric tool. Ear and Hearing, 12(4), 235-241. Han, D . , M o , L . , L i u , H . , Chen, J., & Huang, L . (2006). Threshold estimation in children using auditory steady-state responses to multiple simultaneous stimuli. ORL: Journal for Oto-rhino-laryngology and its Related Specialties, 68(2), 64-68. Harada, J., Aoyagi , M . , Suzuki, T., Kiren, T., & Koike , Y . (1994). A study on the phase spectral analysis of the middle latency reposnse and 40-Hz event-related potential in central nervous system disorders. Acta Otolaryngologica, Supplimental 511, 34-39. Hatton, J., & Stapells, D . R. (2006). Efficiency of single versus multiple stimulus auditory steady-state responses in infants. Paper presented at the Poster accepted for presentation at the Annual meeting of the 2006 American Auditory Society ( A A S ) , Scottsdale, A Z . (March 5-7, 2006). 64 Herdman, A . T., Lins, O. G . , van Roon, P., Stapells, D . R., Scherg, M . , & Picton, T. W . (2002). Intracerebral sources of human auditory steady-state responses. Brain Topography, 15(2), 69-86. Herdman, A . T., & Stapells, D . R. (2001). Thresholds determined using the monotic and dichotic multiple auditory steady-state response technique in normal hearing subjects. Scandanavian Audiology, 30, 41-49. Herdman, A . T., & Stapells, D . R. (2003). Auditory steady-state response thresholds of adults with sensorineural hearing impairments. International Journal of Audiology, 42, 237-248. Hyde, M . L . (1994). The slow vertex potential: Properties and clinical applications. In J. T. Jacobson (Ed.), Principles and Applications in Auditory Evoked Potentials (pp. 179-218). Needham Heights, M A : A l l y n & Bacon. Jerger, J. , Chmiel, R., Frost, J. D . , & Coker, N . (1986). Effect of sleep on the auditory steady-state evoked potential. Ear and Hearing, 7(4), 240-245. John, M . S., Dimitrijevic, A . , & Picton, T. W . (2003). Efficient stimuli for evoking auditory steady-state responses. Ear and Hearing, 24, 406-423. John, M . S., Dimitrijevic, A . , van Roon, P., & Picton, T. W . (2001). Multiple auditory steady-state responses to A M and F M stimuli. Audiology & Neuro-otology, 6, 12-27. John, M . S., Lins, O. G . , Boucher, B . L . , & Picton, T. W . (1998). Multiple auditory steady-state responses ( M A S T E R ) : Stimulus and recording parameters. Audiology, 37, 59-82. John, M . S., & Picton, T. W . (2000). Human auditory steady-state responses to amplitude-modulated tones: phase and latency measurements. Hearing Research, 141, 57-79. 65 John, M . S., Purcell, D . W . , Dimitrijevic, A . , & Picton, T. W . (2002). Advantages and caveats when recording steady-state responses to multiple simultaneous stimuli. Journal of the American Academy of Audiology, 73,246-259. Johnson, T. A . , & Brown, C . J. (2005). Threshold prediction using auditory steady-state response and the tone burst auditory brain stem response: A within-subject comparison. Ear and Hearing, 26, 559-576. Kraus, N . , Ozdamar, O., Hier, D . , & Stein, L . (1982). Auditory middle latency responses (MLRs) in patients with cortical lesions. Electroencephalography and clinical Neurophysiology, 54, 275-287. Kuwada, S., Anderson, J. S., Batra, R., Fitzpatrick, D . C , Teissier, N . , & D'Angelo, W . R. (2002). Sources of the scalp-recorded amplitude-modulation following response. Journal of the American Academy of Audiology, 13, 188-204. Kuwada, S., Batra, R., & Maher, V . L . (1986). Scalp potentials of normal and hearing impaired subjects in response to sinusoidally amplitude-modulated tones. Hearing Research, 21, 179-192. Lev i , E . C , Folsom, R. C , & Dobie, R. A . (1993). Amplitude-modulation following response ( A M F R ) : Effects of modulation rate, carrier frequency, age, and state. Hearing Research, 68, 42-52. Linden, D . , Campbell, K . B . , Hamel, G . , & Picton, T. W . (1985). Human auditory steady-state evoked potentials during sleep. Ear and Hearing, 6(3), 167-174. 66 Lins, O. G . , Picton, P. E . , Picton, T. W . , Champagne, S . C , & Durieux-Smith, A . (1995). Auditory steady-state responses to tones amplitude-modulated at 80-110 Hz . Journal of the Acoustical Society of America, 97(5), 3051-3063. Lins, O. G . , & Picton, T. W . (1995). Auditory steady-state responses to multiple simultaneous stimuli. Electroencephalography and Clinical Neurophysiology, 96, 420-432. Lins, O. G . , Picton, T. W. , Boucher, B . L . , Durieux-Smith, A . , Champagne, S. C , Moran, L . M . , Perez-Abalo, M . C , Martin, V . , & Savio, G . (1996). Frequency-specific audiometry using steady-state responses. Ear and Hearing, 17, 81-96. Luts, H . , Desloovere, C , & Wouters, J. (2006). Cl inical application of dichotic multiple-stimulus auditory steady-state responses in high-risk newborns and young children. Audiology & Neuro-otology, 11(1),24-37. Luts, H . , & Wouters, J. (2005). Comparison of M A S T E R and A U D E R A for measurement of auditory steady-state responses. International Journal of Audiology, 44, 244-253. Makela, J. P., & Hari , R. (1987). Evidence for cortical origin of the 40-Hz auditory evoked response in man. Electroencephalography and clinical Neurophysiology, 66, 539-546. Mauer, G . , & Doring, W . H . (1999). Generators of amplitude modulation following response (AMFR). Paper presented at the 16th meeting of the International Evoked Response Audiometry Study Group, Tromso, Norway. (June, 1999). Mil ford , C . A . , & Birchall , J. P. (1989). Steady-state auditory evoked potentials to amplitude-modulated tones in hearing-impaired subjects. British Journal of Audiology, 23, 137-142. M o , L . , & Stapells, D . R. (submitted). The effect of stimulus duration on the brainstem auditory steady-state response. 67 Moore, B . C . J. (1995). Frequency analysis and masking. In B . C. J. Moore (Ed.), Hearing. Handbook of Perception and Cognition (2nd ed., pp. 161-205). New York: Academic Press. Musiek, F . E . , & Lamb, L . (1994). Central auditory assessment: A n overview. In J. Katz (Ed.), Handbook of Clinical Audiology (4th ed., pp. 197-211). Philadelphia: Lippincott, Wil l iams & Wilkins . Palmer, A . R. (1995). Neural signal processing. In B . C. J. Moore (Ed.), Handbook of Perception and Cognition: Hearing (2nd ed., pp. 75-121). San Diego: Academic Press. Perez-Abalo, M . C , Savio, G . , Torres, A . , Martin, V . , Rodriguez, E . , & Galan, L . (2001). Steady state responses to multiple amplitude-modulated tones: A n optimized method to test frequency-specific thresholds in hearing impaired children and normal-hearing subjects. Ear and Hearing, 22,200-211. Pethe, J., Muhler, R., Siewert, K . , & von Specht, H . (2004). Near-threshold recordings of amplitude modulation following responses ( A M F R ) in children of different ages. International Journal of Audiology, 43'(6), 339-345. Petitot, C , Collet, L . , & Durrant, J. D . (2005). Auditory steady-state responses (ASSR) : Effects of modulation and carrier frequencies. International Journal of Audiology, 44, 567-573. Pickles, J. O. (1986). An Introduction to the Physiology of Hearing (2nd ed.). London: Academic Press. Picton, T., John, M . , Dimitrijevic, A . , & Purcell, D . (2003). Human auditory steady-state responses. International Journal of Audiology, 42, 177-219. 68 Picton, T. W. , Dimitrijevic, A . , John, M . S., & van Roon, P. (2001). The use of phase in the detection of auditory steady-state responses. Clinical Neurophysiology, 112, 1698-1711. Picton, T. W . , Durieux-Smith, A . , Champagne, S. C , Whittingham, J., Moran, L . M . , Giguere, C , & Beauregard, Y . (1998). Objective evaluation of aided thresholds using auditory steady-state responses. Journal of the American Academy of Audiology, 9, 315-331. Picton, T. W . , John, M . S., Dimitrijevic, A . , & Purcell, D . (2003). Human auditory steady-state responses. International Journal of Audiology, 42, 177-219. Picton, T. W . , Linden, R. D . , Hamel, G . , & Maru, J. T. (1983). Aspects of averaging. Seminars in Hearing, 4, 327-341. Picton, T. W . , Skinner, C. R., Champagne, S. C , Kellett, A . J. C , & Maiste, A . C . (1987). Potentials evoked by the sinusoidal modulation of the amplitude or frequency of a tone. Journal of the Acoustical Society of America, 82(1), 165-178. Picton, T. W . , Stapells, D . R., Perrault, N . , Baribeau-Braun, J. , & Stuss, D . T. (1984). Human Event-Related Potentials: Current Perspectives. In R. H . Nodar & C. Barber (Eds.), Evoked Potentials II. Boston: Butterworth Publishers. Picton, T. W. , & Stuss, D . T. (1980). The component structure of the human event-related potentials. In H . H . K . a. L . D . (eds.) (Ed.), Motivation, Motor and Sensory Processes of the Brain, Progress in Brain Research (pp. 17-49): Elsevier/North Holland Biomedical Press. Plourde, G . , & Picton, T. W . (1990). Human auditory steady state responses during general anaesthesia. Anesthesia & Analgesia, 77,460-468. 69 Plourde, G., Stapells, D. R., & Picton, T. W. (1991). The human auditory steady-state potentials. Acta Otolaryngologica Supplement, 491, 153-160. Plourde, G., & Villemure, C. (1996). Comparison of the effects of enflurane/N20 on the 40-Hz auditory steady-state response versus the auditory middle-latency response. Anesthesia and Analgesia, S2, 75-83. Ranee, G., Beer, D. E., Cone-Wesson, B. , Shepherd, R. K. , Dowell, R. C , King, A . M . , Rickards, F. W., & Clark, G. M . (1999). Clinical findings for a group of infants and young children with auditory neuropathy. Ear and Hearing, 20(3), 238-252. Ranee, G., & Rickards, F. W. (2002). Prediction of hearing thresholds in infants using steady-state evoked potentials. Journal of the American Academy of Audiology, 13, 236-245. Ranee, G., Rickards, F. W., Cohen, L. T., De Vidi , S., & Clark, G. M . (1995). The automated prediction of hearing thresholds in sleeping subjects using auditory steady-state evoked potentials. Ear and Hearing, 16, 499-507. Rees, A. , Green, G. G. R., & Kay, R. H . (1986). Steady-state evoked responses to sinusoidally amplitude-modulated sounds recorded in man. Hearing Research, 23, 123-133. Regan, D. (1989). Evoked Potentials and Evoked Magnetic Fields in Science and Medecine. New York: Elsevier. Rickards, F. W., & Clark, G. M . (1984). Steady-state evoked potentials to amplitude-modulated tones. In R. H. Nodar & C. Barber (Eds.), Evoked Potentials II: Boston: Butterworth Publishers. 70 Romani, G . L . , Will iamson, S. J., Kaufman, L . , & Brenner, D . (1982). Characterization of the human auditory cortex by the neuromagnetic method. Experimental Brain Research, 47, 381-393. Ross, B . , Draganova, R., Picton, T. W . , & Pantev, C . (2003). Frequency specificity of 40-Hz auditory steady-state responses. Hearing Research, 186, 57-68. Ross, B . , Roberts, L . E . , & Pantev, C . (2000). A high-precision magnetoencephalographic study of human auditory steady-state responses to amplitude-modulated tones. Journal of the Acoustical Society of America, 108(2), 679-691. Santarelli, R., Mauriz i , M . , Conti, G . , Ottaviani, F. , Paludetti, G . , & Pettorossi, V . E . (1995). Generation of human auditory steady-state responses (SSRs). II: Addit ion of responses to individual stimuli. Hearing Research, 83, 9-18. Savio, G . , Cardenas, J., Perez-Abalo, M . , Gonzales, A . , & Valdes, J. (2001). The low and high frequency auditory steady-state responses mature at different rates. Audiology and Neurootology, 6, 279-287. Schreiner, C. E . , & Urbas, J. V . (1986). Representation of amplitude modulation in the auditory cortex of the cat. I. The anterior auditory field ( A A F ) . Hearing Research, 21, 227-242. Small, S. A . , & Stapells, D . R. (2005a). Multiple auditory steady-state response thresholds to bone-conduction stimuli in adults with normal hearing. Journal of the American Academy ofAudiology, 16, 172-183. Small, S. A . , & Stapells, D . R. (2005b). Normal ipsilateral/contralateral asymmertries in infant bone-conduction multiple auditory steady-state responses. Paper presented at the 71 International Evoked Response Audiometry Study Group ( IERASG) , Havana, Cuba. (June 12-16, 2005) Spydell, J. D . , Pattee, G . , & Goldie, W . D . (1985). The 40 Hertz auditory event-related potential: normal values and effects of lesions. Electroencephalography and Clinical Neurophysiology, 62, 193-202. Stapells, D . R. (2002). Cortical event-related potentials to auditory stimuli. In J. Katz (Ed.), Handbook of Clinical Audiology (5th ed., pp. 378-406). Baltimore: Lippincott, Wil l iams & Wilkins . Stapells, D . R., Galambos, R., Costello, J. A . , & Makeig, S. (1988). Inconsistency of auditory middle latency and steady-state responses in infants. Electroencephalography and Clinical Neurophysiology, 77,289-295. Stapells, D . R., Linden, D . , Suffield, J. B . , Hamel, G . , & Picton, T. W . (1984). Human auditory steady-state potentials. Ear and Hearing, 5(2), 105-113. Stapells, D . R., Makeig, S., & Galambos, R. (1987). Auditory steady-state responses: threshold prediction using phase coherence. Electroencephalography and clinical neurophysiology, 67, 260-270. Stapells, D . R., & Van Maanen, A . (2006). Multiple ASSR thresholds to air-conduction stimuli in young infants with normal hearing. Paper presented at the Newborn Hearing Screening 2006, Cernobbio, Italy. (May 31-June 3, 2006). Stefanatos, G . A . (1993). Frequency modulation analysis in children with Landau-Kleffner syndrome. Annals of the New York Academy of Science, 682,412-414. 72 Stefanatos, G . A . , Foley, C . , Grover, W. , & Doherty, B . (1997). Steady-state auditory evoked responses to pulsed frequency modulations in children. Electroencephalography and clinical Neurophysiology, 104,31-42. Stefanatos, G . A . , G . , G . G . , & Ratcliff, G . G . (1989). Neurophysiological evidence of auditory channel anomalies in developmental dysphasia. Archives of Neurology, 46(%), 871-875. Suzuki, T., & Kobayashi, K . (1984). A n evaluation of 40-Hz event-related potentials in young children. Audiology, 23, 599-604. Suzuki, T., Kobayashi, K . , & Umegaki, Y . (1994). Effect of natural sleep on auditory steady state responses in adult subjects with normal hearing. Audiology, 33, 274-279. Suzuki, Y . (2002). Sources of the scalp-recorded amplitude modulation following response. Journal of the American Academy of Audiology, 13, 188-204. Swanepoel, D . , Schmulian, D . , & Hugo, R. (2004). Establishing normal hearing with the dichotic multiple-frequency auditory steady-state response compared to an auditory brainstem response protocol. Acta Otolaryngologica, 124, 62-68. Szalda, K . , & Burkard, R. (2005). The effects of nembutal anesthesia on the auditory steady-state response (ASSR) from the inferior colliculus and auditory cortex of the chinchilla. Hearing Research, 203(1-2), 32-44. Tachisawa, T. (1997). 40-Hz auditory steady-state response (40-Hz SSR). Brain and Nerve, 49, 1123-1129. Umegaki, Y . (1995). Auditory steady-state response to sinusoidally amplitude-modulated tones second report: investigation of response in sleeping state. Journal of the Oto-rhino-laryngological Society of Japan, 98, 430-441. 73 van der Reijden, C . S., Mens, L . H . , & Snik, A . F. (2005). E E G derivations providing auditory steady-state responses with high signal-to-noise ratios in infants. Ear and Hearing, 26, 299-309. Van Maanen, A . , & Stapells, D . R. (2005). Comparison of multiple auditory steady-state responses (80 versus 40) and slow cortical potentials for threshold estimation in hearing-impaired adults. International Journal of Au diology, 44, 613-624. Vander Werff, K . R., Brown, C . J. , Gienapp, B . A . , & Schmidt Clay, K . M . (2002). Comparison of auditory steady-state response and auditory brainstem response thresholds in children. Journal of the American Academy of Audiology, 75,227-235. Will iamson, S. J., Karron, D . , & Kaufman, L . (1991). Advantages and limitations of magnetic source imaging. Brain Topography, 4(2), 169-180. Wong, W . Y . S., & Stapells, D . R. (2004). Brain stem and cortical mechanisms underlying the binaural masking level difference in humans: A n auditory steady-state response study. Ear and Hearing, 25, 57-67. 74 Appendix A : Amplitude data for all conditions for all subjects 75 Appendix A : Amplitude (nV) data for all conditions for all subjects 14 H z Data M l D2 M2 M4 Subject 2 H z sep 4 H z sep 6 H z sep 2 H z sep 4 H z sep 6 H z sep 2 H z sep 4 H z sep 6 H z sep S2 178 144 156 128 147 137 157 55 50 53 S3 163 101 129 142 124 110 91 73 77 47 S6 108 109 106 136 121 119 107 139 56 39 S9 165 141 168 138 180 156 186 66 117 . 122 S10 171 105 79 101 85 114 144 76 47 40 S l l 195 102 143 211 131 259 132 100 155 183 S12 152 118 88 42 121 166 199 89 34 78 S13 344 246 350 306 204 286 220 57 65 80 S14 149 199 276 234 194 174 148 100 107 197 S15 262 173 202 321 258 405 373 147 133 167 S17 236 273 352 289 210 220 170 249 127 360 S18 179 160 142 165 285 207 150 96 87 48 40 H z Data M l D2 M2 M4 Subject 2 H z sep 4 H z sep 6 H z sep 2 H z sep 4 H z sep 6 H z sep 2 H z sep 4 H z sep 6 H z sep S2 107 80 62 84 92 79 87 21 50 31 S3 153 135 133 137 107 105 96 82 74 76 S6 256 155 222 216 179 186 215 142 88 112 S9 317 291 378 375 293 321 319 177 137 225 S10 336 254 222 175 227 148 151 149 89 135 S l l 398 314 283 322 261 216 266 131 131 137 S12 232 168 262 255 114 132 136 79 90 105 S13 371 331 331 331 299 287 263 167 187 129 S14 294 213 157 236 110 137 173 106 56 54 S15 557 350 341 382 243 263 267 162 227 227 S17 473 427 427 434 309 341 341 210 256 215 S18 419 243 227 285 226 205 236 101 161 120 76 80 Hz Data M l D2 M2 M4 Subject 2 Hz sep 4 Hz sep 6 Hz sep 2 Hz sep 4 Hz sep 6 Hz sep 2 Hz sep 4 Hz sep 6 Hz sep S2 82 75 68 64 76 89 61 27 25 24 S3 316 293 302 211 94 103 154 159 165 128 S6 103 37 24 36 37 3 15 93 58 64 S9 230 240 213 342 116 99 191 131 99 118 S10 178 97 137 94 65 55 57 102 73 86 S l l 219 228 209 209 118 105 116 108 93 102 S12 97 106 104 95 47 35 48 83 69 57 S13 189 202 173 203 179 165 172 126 123 123 S14 104 107 78 136 75 91 67 115 100 89 S15 191 145 118 120 91 87 75 100 80 86 S17 235 224 209 2 3 0 184 119 119 97 120 77 S18 224 142 131 136 96 103 125 73 110 84 Conditions: monotic single (Ml), dichotic two-tone (D2), monotic two-tone (M2), monotic four-tone (M4) 77 Appendix B : Amplitude data for all subjects pooled over separation 78 Appendix B : Amplitude (nV) data for all subjects pooled over separation 14 Hz 40 Hz 80 Hz Subject M l D2 M2 M4 M l D2 M2 M4 M l D2 M2 M4 S2 178 142 147 52 107 75 86 34 82 69 75 25 S3 163 124 108 65 153 135 102 77 316 268 117 150 S6 108 117 115 78 256 197 193 114 103 32 18 71 S9 165 149 174 101 317 348 311 179 230 265 135 116 S10 171 95 114 54 336 217 175 124 178 109 59 87 S l l 195 152 174 146 398 306 247 133 219 215 113 101 S12 152 82 162 67 232 228 127 91 97 101 43 69 S13 344 300 236 67 371 331 283 161 189 192 172 124 S14 149 236 172 134 294 202 140 72 104 107 77 101 S15 262 232 345 149 557 357 257 205 191 127 84 88 S17 236 304 200 245 473 429 330 227 235 221 140 98 S18 179 155 214 77 419 251 222 127 224 136 108 89 Conditions: monotic single ( M l ) , dichotic two-tone (D2), monotic two-tone (M2), monotic four-tone (M4) 79 Appendix C: Relative efficiency data for all subjects pooled over separation 80 Appendix C: Relative efficiency data for all subjects pooled over separation 14 HZ 40 HZ 80 HZ Subject M l D2 M2 M4 M l D2 M2 M4 M l D2 M2 M4 S2 0.93 1.05 1.08 0.55 0.33 0.33 0.37 0.21 0.45 0.54 0.59 0.28 S3 0.85 0.91 0.80 0.68 0.47 0.59 0.45 0.47 1.75 2.10 0.92 1.67 S6 0.56 0.86 0.85 0.81 0.79 0.86 0.84 0.70 0.57 0.25 0.14 0.79 S9 0.86 1.10 1.28 1.06 0.97 1.51 1.35 1.10 1.27 2.07 1.06 1.28 S10 0.89 0.70 0.84 0.57 1.03 0.94 0.76 0.76 0.99 0.86 0.46 0.96 S l l 1.02 1.12 1.28 1.52 1.22 1.33 1.07 0.82 1.21 1.69 0.88 1.12 S12 0.79 0.61 1.19 0.70 0.71 0.99 0.55 0.56 0.54 0.80 0.34 0.77 S13 1.79 2.22- 1.74 0.70 1.14 1.44 1.23 0.99 1.05 1.51 1.35 1.37 S14 0.78 1.74 1.27 1.40 0.90 0.88 0.61 0.44 0.58 0.84 0.61 1.12 S15 1.37 1.71 2.55 1.55 1.71 1.55 1.12 1.26 1.06 1.00 0.66 0.98 S17 1.23 2.25 1.47 2.56 1.45 1.86 1.43 1.39 1.30 1.73 1.10 1.08 S18 0.93 1.15 1.58 0.80 1.28 1.09 0.96 0.78 1.24 1.07 0.85 0.99 Conditions: monotic single ( M l ) , dichotic two-tone (D2), monotic two-tone (M2), monotic four-tone (M4) 81 Appendix D : Pilot studies 82 Appendix D : Pilot studies Pilot study 1: Single vs. multiple stimuli, individual amplitudes (nV) Single Multiple Subject 14 Hz 40 Hz 80 Hz 14 Hz 40 Hz 80 Hz 1 275 330 84 233 175 84 2 283 407 105 45 143 64 3 267 365 199 44 190 74 4 152 256 166 N / A N / A N / A 5 214 439 286 N / A N / A N / A 6 139 428 114 N / A N / A N / A 7 N / A 643 537 N / A N / A N / A Average 221 409 213 107 169 74 Pilot study 2: Dichotic vs. monotic stimuli, individual amplitudes (nV) Dichotic Monotic Subject 14 Hz 40 Hz 80 Hz 14 Hz 40 Hz 80 Hz 1 296 N / A N / A 355 N / A N / A 2 78 N / A N / A 174 N / A N / A 3 350 N / A N / A 219 N / A N / A 4 252 242 70 186 199 87 5 501 227 132 247 172 78 6 108 302 160 122 248 99 Average 264 257 120 217 206 88 Pilot study 3: 2 - 6 H z separations, individual amplitudes (nV) 14 Hz 40 Hz 80 Hz Subject 2 Hz sep 4 Hz sep 6 Hz sep 2 Hz sep 4 Hz sep 6 Hz sep 2 Hz sep 4 Hz sep 6 Hz sep 4 186 119 97 135 171 206 59 129 129 5 277 247 227 308 272 297 238 204 222 6 82 122 93 282 224 223 64 55 72 7 N / A N / A N / A 357 348 318 218 229 252 Average 181.666 162.666 139 270.5 253.75 261 144.75 154.25 168.75 Appendix E : Behavioural Research Ethics Board ( B R E B ) approval 84