Impaired Center-Surround Suppression in Patients with Alzheimer’s Disease

Abstract

Alzheimer’s disease (AD) is often associated with declined visual processing abilities. Here we tested whether the functions of center-surround suppression— a hallmark property in the visual system— are altered by AD. To this end, we recruited three groups of participants (AD, elderly, and young) in a motion direction discrimination task, in which we measured the temporal duration threshold of a drifting Gabor with varying stimulus sizes. We first replicated the phenomena of center-surround suppression that the required duration for discriminating a high contrast grating decreases with increasing stimulus size. We then showed that the magnitudes of suppression varied among the three groups. There was progressive reduction of suppression in the elderly and AD groups compared with the young group. Interestingly, we found that the levels of suppression can predict the severity of dementia in the AD group. Our results suggest that AD is associated with impaired center-surround functions in the visual motion processing pathway.

Keywords

Aging Alzheimer’s disease center-surround suppression cortical inhibition motion perception

INTRODUCTION

Besides deficits in short-term memory, spatial cognition, and other executive functions [1, 2], patients with Alzheimer’s disease (AD) have also been shown to exhibit declines in a wide range of visual perceptual capacities including visual motion processing [3 –7]. For instance, it is reported that AD patients were associated with impaired motion sensitivity and speed perception as comparing to healthy elderly subjects [8, 9]. Notably, AD greatly disrupts their ability to use the radial patterns of optic flow to judge and control the direction of self-motion during spatial navigation [10, 11]. These findings provide behavioral evidence supporting the notion that AD leads to a deficit in the magnocelluar pathway of visual motion processing in the brain [7–9 , 12].

Neurophysiologic evidence of abnormal visual processing in AD, however, has been relatively sparse in the literature. In an event-related electroencephalogram study, Fernandez and colleagues measured visual motion evoked potentials in visual cortex, and found that AD patients with lower motion sensitivity had weaker or absent motion evoked potentials [13]. Another study using functional magnetic resonance imaging has shown that AD patients exhibited reduced activations in area V5, superior parietal lobe, and parietal-occipital cortex, relative to healthy elderly controls [14, 15]. At the same time, an emerging circuit-based perspective posits that AD might be linked with disruptions in the cortico-cortical connections [2, 16]. However, the direct evidence supporting such a linkage is still limited [17].

The current study aims to explore this hypothesis by asking a specific question: Are the properties of center-surround suppression, an index of cortical inhibition [18], altered by the presence of AD? Center-surround suppression is a prominent perceptual phenomenon in human vision and it has two different behavioral measures: one is reduced contrast of a stimulus when it is surrounded by high-contrast stimuli [19 –22]; and the other involves reduced motion sensitivity for larger stimuli [23, 24]. The latter is surprising because larger stimuli are typically expected to recruit more neurons and provide more information for the perceptual decision. Both types of surround suppression are considered to reflect the antagonistic interactions between the center and surround parts of receptive fields of cortical neurons [25 –28], and have been applied to normal aging populations with differing conclusions: it is shown that old adults have greater contrast surround suppression [29] but reduced motion surround suppression [24] as comparing toyoung adults.

In the current study we focused on the motionsurround suppression task and applied it across different population groups including healthy elderly and AD patients. Our question is partially motivated by the finding that in the motion surround suppression task, suppressive functions are impaired by normal aging, such that elderly participants show surprisingly superior motion sensitivity over younger controls due to reduced inhibition in the motion sensitive pathways [23, 24]. Here, we want to build on this interesting phenomenon and compare the properties of center-surround suppression between an AD group, age- and sex-matched healthy elderly group, and young adult group.

MATERIAL AND METHODS

Subjects

The current study recruited three groups of participants: a group of AD patients and two healthy control groups (see Table 1 for the detailed demographics of the three groups). The elderly control group consisted of age- and sex-matched normal elderly subjects, most of whom were the spouse or relatives of AD patients. Another control group was made up of young adults mainly from local universities. All subjects had normal or corrected-to-normal vision and were free of other neurologic, ophthalmologic, or psychiatric illnesses, based on their self-report and visual inspections. Due to technical limitations we did not conduct formal ophthalmologic exams on our participants. All the participants including AD patients (or their relatives) gave informed consent for their participation in the study, which was approved by the local Ethics Committee.

Each AD patient met the National Institute of Neurological and Communicative Disorders and Stroke–Alzheimer’s Disease and Related Disorders Association criteria (NINCDS-ADRDA) for the diagnosis of AD [30]. For all subjects in the current study, we measured a score of dementia level according to the Mini-Mental State Exam (MMSE score) [31], prior to the psychophysical test. All AD patients had a MMSE score of less than 27 and normal subjects greater than 27 (see Table 1).

Apparatus and visual stimulus

Subjects were invited to seat in a dimly lit room, facing a CRT monitor (screen resolution: 800×600; refresh rate: 100 Hz) with a binocular viewing distance of 57 cm. A chinrest was used to constrain head movement and keep the viewing condition stable over the course of the experiment. We used the Psychophysical Toolbox [32, 33] to generate the visual stimulus and control the flow of experiments in the MatLab environment (R2015, The MathWorks, Natick, MA). The visual stimulus displayed on the CRT monitor was luminance defined sinusoidal drifting Gabor patch, similar to the one used in the previous studies [23, 24]. The background luminance was 22.9 cd/m². The grating had a spatial frequency of 1 cycle/deg and drifting at a speed of 2 deg/s. The contrast of the grating was either low (5%) or high (95%). Stimulus contrast was spatially modulated by a two dimensional Gaussian envelope. Stimulus size was defined as two standard deviations (2σ) of the Gaussian envelope, and was 1, 2, 3, or 4 degrees of visual angle. The combination of contrasts and stimulus size gave rise to a total of eight taskconditions.

Experimental protocols

For each task condition, we used a staircase procedure to obtain the duration threshold (minimal presentation time) from each participant. As illustrated in Fig. 1, each trial started after subjects pressed the space button on the keyboard with the left hand. After an initial fixation of 500 ms, a Gabor patch was presented foveally drifting either leftward or rightward, with the direction being randomized from trial to trial. Observers had to indicate the perceived direction by pressing one of the two arrow buttons with their right hand. No feedback was provided. Stimulus duration was varied across trials using a 3-down 1-up staircase, converging onto 79.4% correct response. Each staircase procedure started with default stimulus duration of 200 ms. To achieve efficient convergence for the staircase, the step size was varied, with an initial value of 40 ms, switching to 20 ms after two reversals, and 10 ms after another two reversals. Each staircase was terminated after 10 reversals, which usually took no more than 75 trials in all participants. On average, the number of trials needed in each staircase in each stimulus condition was 54.6±13.2 trials (mean±std) for AD patients, and this was not statistically different from the elderly group (52.6±11.2, p = 0.17) but was modestly more than the young group (44.5±9.4, p < 0.05). The duration threshold was calculated as the mean duration of the final six reversal points.

Data analysis

Data analysis was performed using custom-programmed MatLab scripts. Suppression Index (SI) was calculated at both the high and low stimulus contrast for each subject. SI was defined as the log difference of threshold duration between largest and smallest stimulus sizes:

SI = {log}_{10} (Threshold_4 deg) - {log}_{10} (Threshold_1 deg)

Under this convention, positive SI values indicate spatial suppression, i.e., longer durations required to discriminate a large stimulus. In contrast, negative SI values denote spatial summation, i.e., better performance (shorter duration needed) with increasing stimulus sizes.

We used an unbalanced two-way ANOVA, treating participant group and stimulus size as two factors, to perform the statistical tests on both the raw duration thresholds and SIs in the current study. We regarded an alpha value of 0.05 as being statistically significant and marked it with an asterisk.

RESULTS

Disrupted center-surround suppression in AD

We measured the temporal duration threshold as a function of stimulus size in three groups of participants, shown separately for low contrast (5%, Fig. 2A) and high contrast (95%, Fig. 2B). Clearly, there were remarkable differences between the low and high contrasts. At the low contrast, duration threshold decreased rapidly with increasing stimulus size, and this pattern held for all three groups. An unbalanced two-way ANOVA analysis revealed that there were main effects on the stimulus size (F(3,220) = 71.2, p < 0.001) and the participant group (F(2,220) = 25.9, p < 0.001), without significant interactions between the two (F(6,220) = 1.3, p = 0.27). This means, as the stimulus grew larger in size, it became easier for all participants to discriminate the drifting direction of the Gabor patch. This met our expectation that larger stimulus is often associated with enhanced perception. In addition to size effects, elderly and AD participants in general need longer stimulus duration to judge motion directions, as compared with young participants. This might reflect a general decline in the visual abilities associated with normal aging or other neurologicdisorders.

In contrast, for a drifting Gabor at the high contrast (Fig. 2B), opposite trends were observed. The duration threshold increased with increasing stimulus size. In other words, as the size of the stimulus went larger, it became surprisingly more difficult to discriminate the direction of the drifting Gabor. The increased duration with increasing stimulus size was reported in the previous studies [23] and was regarded as a perceptual correlate of antagonistic interactions between center and surround regions of the receptive fields of cortical neurons. The results of an unbalanced two-way ANOVA indicated that there were significant main effects on both the stimulus size (F(3,220) = 35.2, p < 0.001) and the participant group (F(2,220) = 3.7, p = 0.025), as well as a significant interaction between these two factors (F(6,220) = 13.0, p < 0.001). The interaction resulted in a surprising observation that elderly and AD subjects need shorter stimulus duration to discriminate motion direction than young adults when stimulus size was large (e.g., 4 degrees). Post-hoc t-tests at the high contrast of the largest stimulus size revealed that, while both young-elderly and young-AD differences reached statistical significance (both p < 0.001), there were no significant elderly-AD differences(p = 0.15).

To quantify the magnitudes of center-surround suppression and compare them across the three participant groups, for each participant we computed a suppression index (SI), defined as a log difference of duration thresholds between largest and smallest stimulus size at the high stimulus contrast (see Methods). As summarized in Fig. 3, when the contrast was low, the suppression index was negative, indicating that a spatial summation process was mainly taken place in all three groups. A mixed one-way ANOVA analysis revealed no differences among the three groups (F(2,55) = 0.70, p = 0.5). When the contrast was high, the suppression index became largely positive, indicating the presence of surround suppression. The mean SI was 0.32±0.029 (mean±sem) in the young group, declined to 0.15±0.034 in the elderly, and further deteriorated to 0.03±0.030. A mixed one-way ANOVA showed that the group effect was statistically significant (F(2,55) = 24.2, p < 0.001). Post-hoc unpaired t-test between the elderly and AD groups also revealed a significant difference (t(36) = 2.66, p = 0.012). This mean, aging cannot fully explain the reduced suppression observed in AD patients.

Center-surround suppression in AD was correlated with clinical severity

We reasoned that, if the properties of center-surround suppression are indeed related to the emergence of AD, then the severity levels in AD patients should predict the magnitudes of suppressions to some extent. To test this possibility, we correlated SI values at the high contrast with MMSE score, a clinical indication of the severity of AD measured before the psychophysical experiment (Fig. 4). The results showed that SI was positively correlated with MMSE scores (Pearson correlation, r = 0.49 p = 0.026, df = 18). This means AD patients who scored less in the MMSE assessment (more severe) had weaker center-surround suppressions.

To demonstrate the robustness of this correlation between SI levels and MMSE scores, we did the following outlier-exclusion analysis. Firstly, we applied the outlier criterion of mean±2std on the suppression index, and excluded two subjects (with highest and lowest SIs) from the correlation analysis. The Pearson correlation between SI and MMSE score in the remaining subjects became even stronger (r = 0.58, p = 0.012, df = 16). Secondly, when the same criterion was applied on the MMSE score, one subject with the fewest point was excluded, and the correlation was close to the significant level (r = 0.45, p = 0.053, df = 17). Thirdly, when we apply the same criterion on both parameters and excluded all three outliers, the Pearson correlation on the remaining population fell back into the range of statistical significance (r = 0.50, p = 0.041, df = 15).

DISCUSSION

The current study reported counterintuitive findings that AD patients possess superior visual capacities of motion discrimination when visual stimuli are presented at the high contrast with larger size, as comparing to normal, age-matched elderly subjects and young adults. The better visual performance might reflect reduced center-surround antagonistic interactions, not just a general alteration of visual functions in AD patients, as this enhancement was very specific to larger stimuli of high contrast, but not for stimuli with smaller size and low contrast. Importantly, the magnitude of center-surround suppression in the visual task can predict the clinical assessment of AD severity. We conclude that AD alters the properties of center-surround functions in the visual brain, which goes beyond aging-related cortical neurodegeneration.

Progressive decline of center-surround suppression from young to elderly to AD

Our results of declined center-surround suppression do not reflect a general deterioration of visual processing in the AD brain. We had at least two observations in our data that support this interpretation. Firstly, we found that the mean SI was significantly different between the three groups; but this difference was present only when the contrast was high (Fig. 3). When the contrast was low, mean SI became negative, implying a transition from surround suppression to spatial summation. Importantly, the spatial summation did not differ among three groups. This interaction between SI and contrast levels indicates that the suppression difference we observed was not a general one, but uniquely present in some specific stimulus conditions. Secondly, at the level of raw duration threshold, we found that AD patients took shorter durations to perceive motion directions than young participants (Fig. 2B), and this enhancement was most prominent at the largest stimulus size tested in the current study and present only when the contrast was high.

The superior motion perception in AD was, on the one hand, against conventional wisdom of impaired brain functions in clinical populations, and, on the other hand, not totally unexpected. Using the same paradigm, previous studies have already demonstrated that normal aging can lead to superior motion perception as a result of reductions in the functions of center-surround suppression [24]. In the current study we replicated this phenomenon, and more importantly, we went beyond by showing that the presence of AD further disrupted center-surround functions, and that the magnitudes of center-surround suppression were correlated with AD severity (Fig. 4). The progressive nature of weakened suppression in the three participant groups suggests that these behavioral changes might reflect the progression of cortical pathophysiology underlying the transitions from young to elderly and to AD.

Implications on the neural mechanisms of AD

Our results that center-surround interactions were altered by both AD and normal aging (although to different levels) imply that this function may not be a unique biomarker for AD. However, this does not prevent it from serving as an additional indicator tracking the progress of AD. In addition, since center-surround suppression is considered to reflect the interactions between excitatory center and inhibitory surround of receptive fields of cortical neurons [23, 24], it could be speculated that these behavioral alterations might have roots in the cortical inhibitory functions [18 , 34]. This interpretation is consistent with the contemporary notion that many neurologic disorders including AD might be caused by a disruption of excitation-inhibition balance in the brain [35 –38]. The current study support this notion by showing that both normal aging and AD participants exhibited abnormities in inhibition-based center-surround suppressions comparing to young participants. Indeed, neural fingerprints of center-surround suppression have been found in motion-selective neurons from a hierarchy of cortical areas including striate and extra-striate cortex [26 , 39–41]. It should be noted, though, that there are two forms of surround suppression at the behavioral level, as we outlined in the Introduction, and they might recruit independent neuronal mechanisms [42]. Further studies should be implemented to further identify and characterize the specific sources of these cortical alterations in AD patients, with an ultimate goal of enhancing our understanding of visual dysfunctions and developing effective interventions to improve the quality of life of individuals with AD.

Methodological limitations

One of the major limitations of this study is that participants have not been examined for visual acuity and age-related vision pathologies such as cataract and glaucoma, which are common eye diseases in AD patients [43, 44] and have been associated with visual dysfunctions (for a recent comprehensive review, see [45]). It is therefore possible that the impaired center-surround suppressions in some AD patients in the current study are mediated, at least partially, by the presence of these vision abnormalities. We want to point out that precautions should be taken when considering the potential neurophysiologic underpinnings of the observed visual abnormalities in our data. Future studies are needed to determine whether impaired center-surround suppressions in AD are truly a consequence of disrupted cortical excitation-inhibition balance, or caused by the presence of these degenerative vision diseases.

Footnotes

ACKNOWLEDGMENTS

This work was supported by the Program for Science and Technology Development in Liaocheng City (Funding number: 201606350702). We thank three anonymous reviewers for their constructive criticisms and invaluable suggestions for improving the quality of our manuscript.

Authors’ disclosures available online ().

References

Monacelli

, Cushman

, Kavcic

, Duffy

(2003) Spatial disorientation in Alzheimer’s disease: The remembrance of things passed. Neurology 61, 1491–1497.

von Gunten

, Bouras

, Kovari

, Giannakopoulos

, Hof

(2006) Neural substrates of cognitive and behavioral deficits in atypical Alzheimer’s disease. Brain Res Rev 51, 176–211.

Cronin-Golomb

, Corkin

, Growdon

(1995) Visual dysfunction predicts cognitive deficits in Alzheimer’s disease. Optom Vis Sci 72, 168–176.

Rizzo

, Nawrot

(1998) Perception of movement and shape in Alzheimer’s disease. Brain 121(Pt 12), 2259–2270.

Kavcic

, Duffy

(2003) Attentional dynamics and visual perception: Mechanisms of spatial disorientation in Alzheimer’s disease. Brain 126, 1173–1181.

Paxton

, Peavy

, Jenkins

, Rice

, Heindel

, Salmon

(2007) Deterioration of visual-perceptual organization ability in Alzheimer’s disease. Cortex 43, 967–975.

Velarde

, Perelstein

, Ressmann

, Duffy

(2012) Independent deficits of visual word and motion processing in aging and early Alzheimer’s disease. J Alzheimers Dis 31, 613–621.

Gilmore

, Wenk

, Naylor

, Koss

(1994) Motion perception and Alzheimer’s disease. J Gerontol 49, P52–P57.

Kavcic

, Vaughn

, Duffy

(2011) Distinct visual motion processing impairments in aging and Alzheimer’s disease. Vision Res 51, 386–395.

10.

Tetewsky

, Duffy

(1999) Visual loss and getting lost in Alzheimer’s disease. Neurology 52, 958–965.

11.

Mapstone

, Logan

, Duffy

(2006) Cue integration for the perception and control of self-movement in ageing and Alzheimer’s disease. Brain 129, 2931–2944.

12.

Kuang

, Zhang

(2014) Smelling directions: Olfaction modulates ambiguous visual motion perception. Sci Rep 4, 5796.

13.

Fernandez

, Kavcic

, Duffy

(2007) Neurophysiologic analyses of low- and high-level visual processing in Alzheimer disease. Neurology 68, 2066–2076.

14.

Thiyagesh

, Farrow

, Parks

, Accosta-Mesa

, Young

, Wilkinson

, Hunter

, Woodruff

(2009) The neural basis of visuospatial perception in Alzheimer’s disease and healthy elderly comparison subjects: An fMRI study. Psychiatry Res 172, 109–116.

15.

Kuang

(2016) Is reaction time an index of white matter connectivity during training? Cogn Neurosci doi: 10.1080/17588928.2016.1205575

16.

Festa

, Insler

, Salmon

, Paxton

, Hamilton

, Heindel

(2005) Neocortical disconnectivity disrupts sensory integration in Alzheimer’s disease. Neuropsychology 19, 728–738.

17.

, Zhang

, Liu

, Duan

, Wei

, Zhuo

, Li

, Zhang

, Yu

, Wang

, Jiang

(2016) Distinct changes in functional connectivity in posteromedial cortex subregions during the progress of Alzheimer’s disease. Front Neuroanat 10, 41.

18.

Nassi

, Lomber

, Born

(2013) Corticocortical feedback contributes to surround suppression in V1 of the alert primate. J Neurosci 33, 8504–8517.

19.

Cannon

, Fullenkamp

(1991) Spatial interactions in apparent contrast: Inhibitory effects among grating patterns of different spatial frequencies, spatial positions and orientations. Vision Res 31, 1985–1998.

20.

Petrov

, McKee

(2006) The effect of spatial configuration on surround suppression of contrast sensitivity. J Vis 6, 224–238.

21.

Snowden

, Hammett

(1998) The effects of surround contrast on contrast thresholds, perceived contrast and contrast discrimination. Vision Res 38, 1935–1945.

22.

Zenger-Landolt

, Heeger

(2003) Response suppression in v1 agrees with psychophysics of surround masking. J Neurosci 23, 6884–6893.

23.

Tadin

, Lappin

, Gilroy

, Blake

(2003) Perceptual consequences of centre-surround antagonism in visual motion processing. Nature 424, 312–315.

24.

Betts

, Taylor

, Sekuler

, Bennett

(2005) Aging reduces center-surround antagonism in visual motion processing. Neuron 45, 361–366.

25.

Angelucci

, Bressloff

(2006) Contribution of feedforward, lateral and feedback connections to the classical receptive field center and extra-classical receptive field surround of primate V1 neurons. Prog Brain Res 154, 93–120.

26.

Angelucci

, Bullier

(2003) Reaching beyond the classical receptive field of V1 neurons: Horizontal or feedback axons? J Physiol Paris 97, 141–154.

27.

Born

, Bradley

(2005) Structure and function of visual area MT. Annu Rev Neurosci 28, 157–189.

28.

Schwabe

, Obermayer

, Angelucci

, Bressloff

(2006) The role of feedback in shaping the extra-classical receptive field of cortical neurons: A recurrent network model. J Neurosci 26, 9117–9129.

29.

Karas

, McKendrick

(2009) Aging alters surround modulation of perceived contrast. J Vis 9, 11–19.

30.

McKhann

, Drachman

, Folstein

, Katzman

, Price

, Stadlan

(1984) Clinical diagnosis of Alzheimer’s disease: Report of the NINCDS-ADRDA Work Group under the auspices of Department of Health and Human Services Task Force on Alzheimer’s Disease. Neurology 34, 939–944.

31.

Folstein

, Folstein

, McHugh

(1975) Mini-mental state. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 12, 189–198.

32.

Brainard

(1997) The Psychophysics Toolbox. Spat Vis 10, 433–436.

33.

Pelli

(1997) The VideoToolbox software for visual psychophysics: Transforming numbers into movies. Spat Vis 10, 437–442.

34.

Kuang

(2016) Two polarities of attention in social contexts: From attending-to-others to attending-to-self. Front Psychol 7, 63.

35.

Anticevic

, Cole

, Repovs

, Savic

, Driesen

, Yang

, Cho

, Murray

, Glahn

, Wang

, Krystal

(2013) Connectivity, pharmacology, and computation: Toward a mechanistic understanding of neural system dysfunction in schizophrenia. Front Psychiatry 4, 169.

36.

Isaacson

, Scanziani

(2011) How inhibition shapes cortical activity. Neuron 72, 231–243.

37.

Murray

, Anticevic

, Gancsos

, Ichinose

, Corlett

, Krystal

, Wang

(2014) Linking microcircuit dysfunction to cognitive impairment: Effects of disinhibition associated with schizophrenia in a cortical working memory model. Cereb Cortex 24, 859–872.

38.

Kuang

(2016) Toward a unified social motor cognition theory of understanding mirror-touch synaesthesia. Front Hum Neurosci 10, 246.

39.

Bair

, Cavanaugh

, Movshon

(2003) Time course and time-distance relationships for surround suppression in macaque V1 neurons. J Neurosci 23, 7690–7701.

40.

Eifuku

, Wurtz

(1998) Response to motion in extrastriate area MSTl: Center-surround interactions. J Neurophysiol 80, 282–296.

41.

Sceniak

, Hawken

, Shapley

(2001) Visual spatial characterization of macaque V1 neurons. J Neurophysiol 85, 1873–1887.

42.

Yazdani

, Serrano-Pedraza

, Whittaker

, Trevelyan

, Read

(2015) Two common psychophysical measures of surround suppression reflect independent neuronal mechanisms. J Vis 15, 21.

43.

Bayer

, Ferrari

, Erb

(2002) High occurrence rate of glaucoma among patients with Alzheimer’s disease. Eur Neurol 47, 165–168.

44.

Goldstein

, Muffat

, Cherny

, Moir

, Ericsson

, Huang

, Mavros

, Coccia

, Faget

, Fitch

, Masters

, Tanzi

, Chylack

Jr , Bush

(2003) Cytosolic beta-amyloid deposition and supranuclear cataracts in lenses from people with Alzheimer’s disease. Lancet 361, 1258–1265.

45.

Albers

, Gilmore

, Kaye

, Murphy

, Wingfield

, Bennett

, Boxer

, Buchman

, Cruickshanks

, Devanand

, Duffy

, Gall

, Gates

, Granholm

, Hensch

, Holtzer

, Hyman

, Lin

, McKee

, Morris

, Petersen

, Silbert

, Struble

, Trojanowski

, Verghese

, Wilson

, Xu

, Zhang

(2015) At the interface of sensory and motor dysfunctions and Alzheimer’s disease. Alzheimers Dement 11, 70–98.