«SPATIAL SUPPRESSION IN AGING THE EFFECT OF AGING ON SPATIAL SUPPRESSION By LINDSAY E. FARBER, H.B.Sc. A Thesis Submitted to the School of Graduate ...»
SPATIAL SUPPRESSION IN AGING
THE EFFECT OF AGING ON SPATIAL SUPPRESSION
LINDSAY E. FARBER, H.B.Sc.
Submitted to the School of Graduate Studies
in Partial Fulﬁllment of the Requirements
for the Degree
Doctor of Philosophy
c Copyright by Lindsay E. Farber, December 2015
DOCTOR OF PHILOSOPHY (2015) McMaster University
(Neuroscience Graduate Program) Hamilton, Ontario TITLE: The eﬀect of aging on spatial suppression AUTHOR: Lindsay E. Farber, H.B.Sc. (Western University) SUPERVISORS: Professors Patrick J. Bennett and Allison B. Sekuler NUMBER OF PAGES: xvii, 130 ii Abstract The eﬀect of aging on spatial suppression Lindsay E. Farber Doctor of Philosophy Neuroscience Graduate Program McMaster University The research discussed here examines how normal healthy aging aﬀects spatial suppressive mechanisms in a variety of visual tasks using both static and dynamic stimuli. Prior research has suggested that younger adults demonstrate a center-surround antagonistic pattern in which they show spatial summation at low contrast and spatial suppression at high contrast in brief motion direction discrimination tasks. Older adults have been shown to have reduced spatial suppression at high contrast and this is thought to be related to reduced GABAergic inhibition in the visual cortex. The results obtained from this program of research suggest that age-related changes in optical and neural visual mechanisms do not aﬀect spatiotemporal mechanisms for static stimuli when the target is presented with the mask (embedded masking). However, when the mask appears immediately before (forward masking) or after (backward masking) the target, older adults require more contrast to detect the target (Chapter 2). In addition, spatial suppression is not reduced for older adults in a task with moving stimuli presented at long durations, even with increasing speed (Chapter 3). In Chapter 4, we used static stimuli presented at brief durations to induce a sudden motion onset and found that although there was no signiﬁcant age diﬀerence in spatial suppression, there was a trend showing reduced levels of spatial suppression in older adults. These results taken together suggest that inhibitory neural mechanisms in the visual cortex may mediate spatial iii suppression for brieﬂy presented stimuli only.
iv Acknowledgements I could not have completed my doctoral dissertation without the unwavering support I received from a fantastic team of people. Though I can’t acknowledge every individual who provided either professional or personal support during my time as a Ph.D. candidate, I’d like to highlight a few of the most signiﬁcant contributors to my success in reaching this milestone.
First, I’m so grateful to my graduate supervisors, Dr. Patrick J. Bennett and Dr. Allison B. Sekuler, for their invaluable guidance, scholarly input, and consistent encouragement, which I was so lucky to receive throughout my time spent in their lab. I’d also like to thank my third Ph.D. committee member, Dr. Bruce K. Christensen, the Director of the MiNDS program, Dr.
Kathy M. Murphy, and the MiNDS coordinator, Sandra Murphy. Kathy and Sandra, the unique roles that you’ve played on this journey have meant a great deal to me.
I wouldn’t be where I am today—academically or otherwise—without the support of my research lab manager, Donna Waxman, who not only helped me collect important data but also strengthened my stamina with her kind words over the years. My Vision Lab colleagues deserve mention for the warmth they extended to me; I learned a lot from them through our personal and scholarly interactions. I cannot thank our senior participants enough for spending countless hours in our lab, contributing to our research projects.
I’d be remiss if I didn’t thank my boss, Dr. Jonathan A. Goler, for his steadfast belief in me, his counsel, and his understanding throughout my time at Moneykey, where I have had the opportunity to work while simultaneously ﬁnalizing my thesis.
Finally, I owe so much to my parents, Rachel and Glenn Farber, who cheered me on at every stage of my academic career, longing to see me realize this dream. I also thank my sister, Stacey Farber, for her motivational pep talks and my grandparents, Dr. Robert Farber, Sally Kert, the late Shaynka
2.1 Detection thresholds replotted from Figure 3 in Saarela and Herzog (2008). Time course of iso-orientation contrast masking for subject TS. Negative SOAs indicate that the target onset preceded the mask onset. The dashed horizontal lines show the control detection thresholds, measured with the target (no-mask). 23 2.2 (a) The horizontally-oriented target and mask stimuli. The central target Gabor’s Gaussian envelope had a standard deviation of 0.25 deg. Three diﬀerent mask types were used in the experiment: (b) a central mask (outer diameter = 1 deg), (c) a surround mask (inner diameter = 1 deg; outer diameter = 8.42 deg) and (d) a combination of both masks (outer diameter = 8.42 deg). 28
2.3 Procedure used in Experiments 1 and 3. Targets were presented in one of two sequential intervals (each 1040 ms); subjects indicated which of the two intervals contained the central target Gabor. In this example, the target (40 ms duration) is presented in interval 1, the mask type is a combination mask (100 ms duration), and the SOA is -40 ms. A central ﬁxation point ﬂickered for 500 ms before the ﬁrst stimulus interval. After the 500 ms ISI, the second stimulus was displayed.............. 30
2.5 Masking thresholds plotted as a function of SOA in ms for younger (dashed lines with square data points) and older (solid lines with circular data points) observers. No-mask detection thresholds for both age groups are shown as single data points at the zero SOA time point. Error bars represent ±1 SEM......... 34
2.7 Target and mask detection thresholds for 9 younger (white bars) and 9 older (grey bars) observers. Error bars represent ±1 SEM. 40
2.8 Masking thresholds plotted as a function of mask contrast for younger (dashed lines with circular data points) and older (solid lines with square data points) observers. The horizontal lines indicate thresholds in the no-mask condition. Error bars represent ±1 SEM. (a) Embedded central mask condition. (b) Embedded combination mask condition. (c) Backward combination mask condition. (d) Forward combination mask condition....... 45 xii
2.9 Normalized thresholds plotted as a function of normalized mask contrast for younger (dashed lines with circular data points) and older (solid lines with square data points) observers. Thresholds were normalized by dividing thresholds obtained with masks by the threshold obtained in the no-mask condition. Mask contrast was normalized by dividing mask contrast by the mask detection thresholds measured in Experiment 2. Error bars represent ±1 SEM. (a) Embedded central mask condition. (b) Embedded combination mask condition. (c) Backward combination mask condition. (d) Forward combination mask condition....... 46
3.1 Replotted from Figure 2 in van der Smagt et al. (2010). PSEs (speeds of the test stimulus, (38% contrast, no surround) when matched to that of the reference stimulus) for all contrast/size combinations, averaged across 9 observers. Error bars depict ±1 SEM. The horizontal dashed line represents the reference speed (1 cps)............................ 67 3.2 (a) Example of one trial from the detection experiment. In this example the ﬁrst interval was blank, followed by an ISI and the second interval containing the stimulus. (b) Example of one trial from the perceived speed experiment. The reference stimulus was always presented in interval 1 and the test stimulus was always presented in interval 2. The stimulus intervals were separated by an ISI........................ 71
3.3 Experiment 1 PSEs for all contrast/size combinations at the 1 cps reference speed for 14 younger observers and 13 older observers. The dashed lines indicate a PSE of 1.......... 76
3.4 Experiment 2 PSEs for all contrast/size combinations at the 4 cps reference speed for 13 younger observers and 10 older observers. The dashed lines indicate a PSE of 4.......... 79 xiii
4.1 Figure replotted from Figure 2B in Churan et al. (2009). Normalized thresholds were calculated to make the trends for each subject independent of their individual performance in motion discrimination. Normalized thresholds (Tnorm = (T - Tmin) / (Tmax - Tmin), where Tmin and Tmax represent the minimal and maximal thresholds obtained from each subject on any condition in Churan et al.’s Experiment 1) were calculated for each of the four subjects. The average of these normalized thresholds are shown in this ﬁgure. Low contrast gratings (1.5% contrast) are shown as white bars and high contrast gratings (98% contrast) are shown as gray bars. There were six stimulus size conditions (5.3, 7.9, 10.5, 13.2, 15.8, and 18.5 deg). Performance improved for low-contrast gratings (thresholds decreased) and worsened for high-contrast gratings (thresholds increased) as stimulus size increased. Error bars depict ±1 SEM....... 95 4.2 (a) Example of one trial from the detection experiment. In this example the ﬁrst interval is blank, followed by an ISI and the second interval containing the stimulus. (b) Example of one trial from the phase step experiment. In this example the ﬁrst interval contains the medium-sized, high-contrast Gabor, followed by an ISI and the second interval containing the mediumsized, high-contrast Gabor with its phase shifted rightwards.. 99
4.3 Mean detection thresholds for 14 younger and 12 older subjects in Experiment 1 for a medium-sized (3.65 deg) Gabor patch.
Error bars represent ±1 SEM................... 102
4.4 Results from the phase shift task in Experiment 1. Mean performance of 14 younger subjects and 12 older subjects for low contrast (white bars) and high contrast (gray bars) Gabor stimuli at diﬀerent stimulus sizes. Error bars represent ±1 SEM.. 104
4.6 Results from the phase shift task in Experiment 2. Mean performance of 10 younger subjects and 12 older subjects for low contrast (white bars) and high contrast (gray bars) Gabor stimuli at diﬀerent stimulus sizes. Error bars represent ±1 SEM.. 108
2.1 Experiment, number of subjects, age, decimal near acuity, decimal far acuity, MMSE and MoCA scores for older subjects.
Values in parentheses are standard deviations.......... 26
3.1 Experiment, number of subjects, age, decimal near acuity, decimal far acuity, MMSE and MoCA scores for older subjects.
Values in parentheses are standard deviations.......... 70
4.1 Experiment, number of subjects, age, decimal near acuity, decimal far acuity, MMSE and MoCA scores for older subjects.
Values in parentheses are standard deviations.......... 97 xviii List of Abbreviations 2-IFC two-interval forced-choice ANOVA ANalysis Of VAriance cd/m2 Candelas per square meter CRT cathode-ray tube CFQ Cognitive Failures Questionnaire cm centimeters cpd cycles-per-degree cps cycles-per-second deg degrees dps degrees-per-second ETDRS Early Treatment Diabetic Retinopathy Study FARS Fatality Analysis Reporting System fps frames-per-second GABA γ-aminobutryic acid
GAD glutamic acid decarboxylase Hz Hertz ISI inter-stimulus interval M Mean MCI mild cognitive impairment MDD major depressive disorder MEGA-PRESS MEscher-GArwood Point REsolved SpectroScopy MMSE Mini-Mental State Examination MOA motion onset asynchrony MoCA Montreal Cognitive Assessment MREB McMaster Research Ethics Board MRS magnetic resonance spectroscopy ms milliseconds MT middle temporal MTA motion termination asynchrony PSE point-of-subjective equality s seconds SCZ schizophrenia SEM Standard Error of the Mean SOA stimulus onset asynchrony
The main goal of this dissertation is to investigate eﬀects of normal healthy aging on spatial suppression in the visual system. The research presented here is in the form of a sandwich thesis with Chapters 2, 3, and 4 written in journal article format. These chapters are manuscripts that are being prepared for journal submission. Chapter 1 will set the general context for the experiments. Background information speciﬁc to each experimental series is provided in the introduction section of its respective chapter. Chapter 5 reviews the ﬁndings presented in this dissertation in context with the literature and proposes ideas for future experiments.
After I wrote an initial draft, my dissertation was revised collaboratively with my supervisors Patrick J. Bennett (PJB) and Allison B. Sekuler (ABS).
As primary author, I oversaw all aspects of the research presented here. All experimental programming for Chapters 2, 3 and 4 was done by PJB and myself. Donna Waxman, our research manager, collected the data for all experiments. I was responsible for the data analysis in all chapters.
The research here was supported by the Canada Research Chairs program, a grant from the Canadian Institutes of Health Research (CIHR) to PJB and ABS, the CIHR Strategic Training Grant on Communication and Social Interaction in Healthy Aging in Masters Year 1 and PhD Year 1 (PJB and ABS were mentors, and I was a trainee), and an Ontario Graduate Scholarship (OGS), which I was granted in Year 4 of my Ph.D.
Canada’s aging population (≥ 65 years of age) has been steadily growing and is projected to increase from 15.3% of the population as of 2013 to approximately 24% by the year 2043 (Statistics Canada, 2014). This demographic trend will have a dramatic eﬀect on every aspect of life in Canada including healthcare, the workforce, transportation and communication. Due to increasing longevity (Statistics Canada, 2012), a large segment of society will have to cope with the sensory, motor, and cognitive deﬁcits that typically accompany normal healthy aging for an increasingly greater portion of their lives.