Sharon Gilaie-Dotan

We have recently found that during shallow (“naturalistic”/incidental) encoding larger images are better remembered. However, it is unclear whether this reflects a genuine influence of image size on mnemonic processes or an epiphenomenal consequence of its influence on perception, and whether this effect robustly generalizes to deeper encoding. Here, we tested these in three experiments (n = 189) differing in encoding depth: naturalistic (freely viewing the images without being informed a memory test will follow), intentional memorization (freely viewing while memorizing the images since a memory test was to follow), or perceptual encoding (reporting for each image whether there was a person in it and whether it was taken indoors or outdoors, without being informed a memory test will follow). During encoding in the perceptual experiment, participants viewed 160 images of different sizes (3°-15.5°) followed by an old-new recognition memory test of 320 images (half “old”). While perceptual measures were at ceiling, memory improved with image size, presenting a dissociation between perception and memory and indicating that the effect of image size on memory is not a mere reflection of limited perception for smaller images. Additionally, the effect of image size on memory generalized to all encoding tasks such that memory improved for larger images (as evident by accuracy and d’), regardless of average memory performance increasing with encoding depth. These findings suggest that image size operates as a genuine sensory-based modulator of memory encoding, beyond perceptual sensitivity, pointing to a broader role of low-level visual properties in shaping memory representations.

Simmons, Nelson, and Simonsohn (2011) proposed disclosure-based practices — a priori power analyses, full reporting of conditions and exclusions, results reported with and without exclusions applied — as a remedy for the replication crisis. Following these practices explicitly, our lab attempted to publish an online replication and extension of our earlier PNAS findings on image size and memory, expecting that doing so would make publication straightforward. It did not. Nine submissions over 2.5 years were unsuccessful (eight rejections and one withdrawal), largely on the grounds that a transparent, successful replication lacked sufficient theoretical novelty to warrant publication on its own; it was accepted only after this framing was substantially minimized in favor of additional analyses and an alternative theoretical framework. Here, I reflect on this experience, raise open questions about whether the field's evaluation of replications has kept pace with its stated commitment to them, and offer recommendations for future practices.

Visual memory is known to be influenced by objective image properties, including image size and contrast, as well as visual category. However, the influence of illusory perceptual experience on memory remains less understood. Here we investigated whether face pareidolia, a robust illusory phenomenon where certain stimuli elicit a face-like percept despite not being real faces, influences memory. While we assumed illusory faces would be remembered better than matched control images since they are surprising, interesting, draw more attention and may be associated with longer observation times, old-new recognition memory sensitivity was not consistently influenced by the face illusion. Surprisingly, decision criterion was more liberal for illusory faces than for other categories, and this effect scaled with illusory strength such that participants were more likely to report having seen illusory faces for stronger illusory face percepts. In a two-alternative-forced-choice recognition memory paradigm that minimizes criterion influence, we found that the tendency to report having seen illusory faces is present even for unseen images. Further analyses, involving confidence ratings indicated our findings probably reflect a heightened sense of familiarity for illusory faces and not genuine false memory. Our results suggest that during retrieval, illusory faces influence decisions independently of the actual memory trace.

Visual memory is known to be influenced by objective image properties, including image size and contrast as well as visual category. However, the influence of illusory perceptual experience on memory remains less understood. Here we investigated whether face pareidolia, a robust illusory phenomenon where certain stimuli elicit a face-like percept despite not being real faces, influences memory. While we assumed illusory faces would be remembered better than matched control images since they are surprising, interesting, draw more attention, and may be associated with longer observation times, oldnew recognition memory sensitivity was not consistently influenced by the face illusion. Surprisingly, decision criterion was more liberal for illusory faces than for other categories, and this effect scaled with illusory strength such that participants were more likely to report having seen illusory faces for stronger illusory face percepts. In a 2AFC recognition memory paradigm that minimizes criterion influence we found that the tendency to report having seen illusory faces is present even for unseen images. Further analyses involving confidence ratings indicated our findings likely reflect heightened sense of familiarity for illusory faces and not genuine false memory. Our results suggest that during retrieval illusory faces influence decisions independently of the actual memory trace.

Multiple findings suggest that human visual perception exhibits extensive size invariance (i.e., insensitivity to retinal image size changes) for multiple object categories. This is often investigated when retinal image size changes following real-world image size changes, but viewing distance is kept constant, a situation that does not reflect daily vision where objects retain their real-world size and their image size on the retina changes as a function of viewing distance. Here, we measured face discrimination performance to parametrically growing face differences and assessed size invariance by examining whether performance was sensitive to retinal image size changes employing multiple methods (e.g., human performance deviation from that of the ideal observer). Importantly, retinal image size changes were applied either naturalistically (when faces kept their world size but were more distant) or lab-like (where faces changed their world size while viewing distance was constant). We found that (i) the difference detection threshold (DT) was not influenced by retinal size changes or viewing distances we manipulated, (ii) face discrimination was significantly worse for smaller retinal image sizes (for both naturalistic and for lab-like conditions), (iii) worse face discrimination performance at smaller retinal image sizes were significantly associated between naturalistic and lab-like conditions. Our results indicate that face discrimination shows imperfect invariance to retinal size and, importantly, this appears independent of viewing distance. Since face perception may rely on face-specific mechanisms, future work should examine whether these results generalize to additional object categories.

Sensory visual degradation is one of the most well-known implications of aging. Theories on compensation mechanisms in aging suggest that with sensory degradations, processing shifts to prioritize more preserved non-sensory sources of information such as semantic and contextual information. Therefore, it is unclear whether image size changes (evident in profound sensory changes to the retinal image) that significantly affect memory in young adults, would also affect memory in older ages. Here we directly tested this in a large cohort (n=294) covering a continuous age range (18–92 years). Participants freely viewed 160 3-15.5 visual degree sized pictures (of faces, people, indoor and outdoor scenes) without being notified about the memory aspect of the study (“naturalistic encoding”). Afterwards they were given a surprise old-new recognition memory test on these and on 160 additional new images. Image memory was not affected by age (as evident in accuracy, memory sensitivity (dprime), decision criterion and in area under the ROC curve (AUC)). Furthermore, larger images were better remembered across all age groups, and the extent of the image size-related memory benefit was also not affected by age. The findings suggest that even in older age, with sensory senescence, visual (sensory) information affects image memory during naturalist encoding to the same extent as in younger age. These findings may have important implications for understanding memory processes and for developing methods to improve information accessibility and interface design for older adults.

Multiple findings suggest that human visual perception exhibit extensive size invariance (i.e., insensitivity to retinal image size changes) for multiple object categories. This is often investigated when retinal image size changes following real world image size changes but viewing distance is kept constant, a situation that does not reflect daily vision where objects retain their real-world size and their image size on the retina changes as a function of viewing distance. Here we measured face discrimination performance to parametrically growing face differences, and assessed size invariance by examining whether performance was sensitive to retinal image size changes employing multiple methods (e.g. human performance deviation from that of the ideal observer). Importantly, retinal image size changes were applied either naturalistically (when faces kept their world size but were more distant) or lab-like (where faces changed their world size while viewing distance was constant). We found that (i) the difference detection threshold (DT) was not influenced by retinal size changes or viewing distances we manipulated, (ii) face discrimination was significantly worse for smaller retinal image sizes (for both naturalistic and for lab-like conditions), (iii) worse face discrimination performance at smaller retinal image sizes were significantly associated between naturalistic and lab-like conditions. Our results indicate that face discrimination shows imperfect invariance to retinal size and importantly, this appears independent of viewing distance. Since face perception is may rely on face-specific mechanisms, future work should examine whether these results generalize to additional object categories.

We have recently found that people remember bigger images more than smaller images during naturalistic-like encoding. Since these findings were established with accuracy levels which may be confounded by response biases, and since memory performance was apparently low, here we addressed these issues. In a new set of experiments (n=199), we estimated accuracy to test replication of our earlier findings, dprime memory-sensitivity that accounts for potential response biases (criterion), and RT-based ROC measures. Our findings, consistent across experiments, that overall memory as estimated by dprime and ROC-based analyses was significantly above chance, indicate that information was encoded and learnt. Furthermore, larger images (in the 3 to15.5 degrees range) were better remembered as evident in dprime and accuracy. Memory for 1.5 degrees images did not differ from that of 3degrees images and memory for images bigger than 15.5 degrees dropped, likely reflecting limitations of display size. Memorability also increased with image size. Our findings provide converging measures supporting the idea that image size plays an important role in image memory during naturalistic encoding for images in the size range of 3 to 15.5 degrees. Our study also emphasizes the importance of testing an effect across multiple measures and methods.

Some estimates suggest that one in seven good readers and the majority of children with reading difficulties suffer from oculomotor dysfunction (OMD), an umbrella term for abnormalities in comfortable and accurate fixations, pursuits, and saccades. However, national vision evaluation programs worldwide are often limited to distance visual acuity (dVA), not testing for OMD despite its high prevalence and the ease of detecting it in brief optometric evaluations. We hypothesized that reading acquisition is dependent on good oculomotor functions, and therefore inadequate oculomotor control will be associated with reading difficulties. We retrospectively examined and compared oculomotor evaluations (using DEM and NSUCO) and reading assessments (using standardized national reading norms) of a normative class (28 first graders (6–7 yr. olds)) that were independently obtained while blind to the other assessment. Better oculomotor performance as estimated by DEM was associated with better reading performance, and almost a third (29.6%) of the children were categorized by DEM as having OMD-related difficulties. Control analysis revealed dVA was not positively associated with reading performance. Linear regression analyses further corroborated these findings. Since this study is based on a small cohort and since there are studies suggesting that DEM may actually reflect visual processing speed or cognitive factors rather than oculomotor function, replications are needed to substantiate the direct contribution of oculomotor functions to reading acquisition. Young children struggling with reading may benefit from a comprehensive visual evaluation, including oculomotor testing, to provide a more thorough assessment of their learning-related difficulties.

It is unclear how exposure time contributes to image memory during naturalistic visual behavior since most perceptual and memory investigations involve instructed experimental tasks that modulate resource allocation and affect processing from very early stages. We have recently assumed that during naturalistic encoding, without task-based modulations, more physically salient images register better in memory since they entail higher recruitment of early visual system processing leading to better quality signals passing downstream. After finding that spatial image saliency positively contributes to memory (bigger and higher-contrast images are better remembered), here we tested whether during naturalistic encoding temporal image saliency contributes to memory, assuming that images viewed for longer will be better remembered. In a set of experiments (n=182, 250-2000ms exposure time) we found that images viewed for longer (2000ms) were remembered better (13% higher accuracy, 1.43 times more) than images presented for shorter times (250ms), and that this effect was independent of image size effect on memory. These findings suggest that temporal aspects of visual input, much like spatial aspects, play a significant and independent role in memory formation under naturalistic conditions. This underscores the importance of considering exposure time in models of visual memory that aim to reflect real-world behavior.