Perceptual effects of optogenetic stimulation of inferior temporal cortex: hallucination or distortion?


 Local artificial stimulation in high level visual areas of the brain induces complex perceptual events. Anecdotal descriptions of these events fall into two broad categories: ‘hallucinations’, which add a consistent and specific pictorial element to the contents of perception, and ‘distortions’, which transform the ongoing visual perception. Distortions are not pictorially consistent as they vary based on the visual input. Systematic description of such characteristics of stimulation-induced perceptual events is a necessary step for understanding how neural activity gives rise to perception, and is also critical for development of visual prosthetic devices. Here, we implanted arrays of LEDs over inferior temporal cortices of macaque monkeys and trained them to detect and report short optogenetic impulses delivered to their cortices. In a series of experiments, we observed that the ability to detect cortical stimulation highly depends on the choice of images presented to the eyes and that detection of cortical stimulation is most difficult when the animal fixates on a blank screen. Local stimulation of object selective parts of the visual cortex is shown here to induce perceptual events that are easy to detect and contain a strong distortive component. The causal contribution of inferior temporal neurons to perception does not seem to be strongly predetermined, as it depends on the state of vision. These findings also open the door to expanding the scope of visual prosthetics beyond the primary visual cortex.

Perturbation of neural activity in the visual system alters visual perception. Understanding 31 the nature of the perceptual events induced by neural perturbations is essential for bridging the 32 causal gap between neuronal activity and vision as a behavior. This knowledge is crucial for 33 identifying the neural underpinnings of visual hallucinations in psychiatric disease and 34 developing effective visual prosthetics for patients with severe visual impairment. 35 Verbal reports of human patients describe two different types of perceptual events induced  Afraz et al., 2015). Stimulation of face-selective parts of IT cortex is shown to strongly 77 affect match-to-sample performance for faces but not other stimuli (Moeller et al., 2017), a result 78 that is suggestive of face-specific distortions, but can be explained by face hallucinations as well 79 because a hallucinatory face may interact with the match-to-sample task more for faces than the 80 other stimuli. While these studies reveal specific perceptual changes resulting from artificial 81 perturbation of the neural activity, they remain mostly agnostic with respect to the hallucinatory 82 versus distortive nature of those changes.

83
In this study, we designed a novel psychophysical task to systematically investigate the  We trained the animals to behaviorally detect a short optogenetic stimulation impulse delivered 95 to their IT cortex while fixating at images of various objects and scenes (Figure 1.b). In each 96 trial, following fixation, an image was displayed on the screen for 1 s. In half of the trials, 97 randomly selected, a 200 ms illumination impulse was delivered to IT cortex halfway through 98 the image presentation, and the animal was rewarded for correctly identifying whether the trial 99 did or did not contain cortical stimulation. The image content was independent of whether brain 100 stimulation would or would not occur and the subjects' exclusive behavioral task was to detect if 101 brain stimulation occurred in a given trial. We found this approach produced robust and large 102 behavioral effects. Here we present the results of a series of experiments deploying this tactic, 103 beginning with an experiment designed to determine whether optogenetic stimulation of IT 104 cortex evokes a detectable visual event and culminating with a systematic test of the 105 hallucination versus distortion hypotheses.        In order to tease apart these two interpretations, we tested how attenuation of visibility of 187 screen images affect detection of the cortical event. The animals performed the stimulation-188 detection task while fixating on randomly presented images of five objects at four visibility 189 levels in addition to a no image condition ( Figure S1). Visibility was degraded by reducing the    Array at the beginning of each trial and trial delivery was paused if the temperature on the LED 280 die rose more than 3° C above the baseline temperature, and restarted once they were less than 1° 281 C above the baseline. 3° C at the LED die translates to approximately 0.5° C temperature change 282 on the cortical surface; this temperature management regime is detailed in Rajalingham et al.   Sp. started training with all 22 images, but we eventually reduced the number of images to 1 and 329 slowly reintroduced the full training set like in Ph. Then, in both monkeys we reduced the 330 number of activated LEDs to one, and illumination power to 4.5 mW in Ph and 9.1 mW in Sp. 331 We introduced catch trials to Ph. after 17 sessions at an initial rate of 5% of all trials, then after 332 23 sessions increased the rate to 10% of all trials which continued for the rest of training. Catch     Figure S3). Where i is the trial number. x is illumination power, α, β, β1, …, βn are the fit coefficients. 420 For each trial, the λ that matches the image index is assigned 1, and the rest are assigned 0.        Competing interests: 568 The authors declare that there is no conflict of interest.

569
Data and material availability: 570 The data and material that support the findings of this study are available on request from the 571 corresponding author R.A.  implies that cortical stimulation adds a speci c visual element (e.g. a monkey face) to the contents of visual perception independent of the external visual stimulus. Alternatively, the distortion hypothesis (right column) assumes that stimulation-induced perceptual events highly depend on the visual input, predicting a unique event for each visual stimulus viewed. b) Behavioral task: in each behavioral trial following xation an image was displayed on the screen for 1 s. In half of the trials, randomly selected, a 200 ms illumination impulse was delivered to IT cortex halfway through image presentation. The animal was rewarded for correctly identifying whether the trial did or did not contain cortical stimulation by looking at one of the two subsequently presented targets at the end of trial. c) Schematic illustration of the procedure for chronic optogenetic stimulation of IT cortex. Left, Injection of Adeno-associated virus (serotype 5) engineered to express the excitatory opsin C1V1. Right, Opto-Array implantation: in a separate surgery, we visually con rmed the expression of the excitatory virus and implanted an Opto-Array over the expression zone. We also implanted another array in the corresponding region of the opposite hemisphere where no virus was injected (control site, not shown). d) Behavioral performance as a function of session number during the training phase. The y-axis indicates the proportion of the trials reported as stimulated. Red, blue and yellow colors represent data from the stimulation, non-stimulation, and catch trials respectively. Error bars represent bootstrapped 95% con dence intervals. The difference between stimulated and nonstimulated trials became signi cant at session 4 (arrow) and remained so through the training. Fluctuations of performance in time represent usage of different visual stimuli and stimulation intensities throughout the training. No signi cant difference was found between the catch and non-stimulation trials. The violin plots on the right side illustrate the mean and bootstrapped 95% con dence interval of stimulation report rate for each trial type in the last 3 sessions, between the dashed lines.

Figure 2
Stimulation detection performance is modulated by visual input, cortical location, and illumination power. a) left, detection pro le: the behavioral performance (d') on the cortical stimulation detection task for 40 images. The black dots represent d' for each image and the violin plots represent bootstrapped 95% con dence intervals. Right, permutation test: the blue line indicates the standard deviation of d's across images, and the red histogram represents results from a permutation test with 10,000 times randomly assigned images on trials revealing the statistical signi cance of the effect of image on performance. b) left, correlation between detection pro les within each cortical stimulation site and between them. The violin plots represent 95% con dence intervals of the bootstrapped distribution of the correlations with 10,000 resamples, and the horizontal lines indicate their medians. Right, permutation test: the blue line indicates the observed correlation between the sites. The red histogram represents results from a permutation test with 10,000 times random assignment of stimulation condition over the trials. This shows that the correlation of detection pro le patterns between the sites is signi cantly lower than the null distribution. c) left, detection performance (d'), as a function of illumination power. Each line represents data from 1 image (5 images in total including the no image condition). Right, permutation test, the standard deviation of the coe cients for each image, derived from tting of the psychometric curves. The blue line indicates the observed value, and the red distribution represents the null distribution generated by 10,000 times randomly assigning the image indexes to the trials. This con rms the coe cients are signi cantly different from each other.

Figure 3
Stimulation detection performance is modulated by image visibility. a) prediction: in case of hallucination, decreasing the visibility of the screen image should either not affect detectability of cortical stimulation (yellow line) or help it by decreasing background clutter (red line). In case of distortion, increasing the visibility should increase the detection performance (blue line), since the perceptual effect is a function of the visual input. b) observation: the x-axis represents 4 levels of image visibility and the gray background, used in experiment 3. The y-axis is the detection performance (d') on the cortical stimulation detection task. The thin lines represent data from 5 different images and the thick line illustrates the overall averages. Error bars represent 95% con dence intervals. There is a signi cant correlation between the image visibility and performance (r = 0.7). The p-values for pairwise comparisons are from post-hoc tests of ANOVA (Benjamini-Hochberg corrected).

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. Supplementary.pdf