• Bai, Ongchoco, Weisman, Schille-Hudson, Ihm, Taves, Luhrmann, & Scholl
• Bocheva & Scholl
• Erdogan & Scholl
• Ji & Scholl
• Jones, Bai, Scholl, & Awh
• McDougle, Bai, Scholl, & Sun
• Verosky & Scholl

The input to visual perception is a continuous wash of light, but what we see are structured scenes of discrete objects. Discrete structure is so fundamental that the visual system may effectively create it in its absence. A particularly striking example occurs in the phenomenon of 'scaffolded attention': when staring at a regular grid of squares (as in graph paper, or floor tiles), many people report spontaneously seeing the squares as grouped into a shifting array of structured patterns (e.g. a block-letter E). (This phenomenon does not occur without the grid, which serves as a 'scaffold' for attention to group subsets of squares.) Might such 'everyday hallucinations' of structure relate to higher-order experiences and beliefs? To find out, we explored its relationship to some of the most profound and life-changing experiences in many people's lives: experiences of supernatural presences (as when seeing a vision of a deity). Observers viewed a uniform grid of squares, and were then asked if they saw the squares as grouped into any shapes or patterns, after which they deliberately tried to see such patterns. The same observers then completed a nuanced measure of supernatural experiences (the Spiritual Events scale). Across a large sample (n=1000), these two forms of experience were related: people who experienced 'supernatural presence' were considerably more likely to spontaneously experience scaffolded attention. And yet, these people were actually *less* able to do so when explicitly instructed -- suggesting that the connection may be specific to *spontaneous* experiences (and cannot be due merely to suggestibility). Crucially, these links could not be explained by general degrees of religiosity, or by demographic factors such as education level. These findings suggest a robust link between two types of experiences that may seem maximally distant: perceiving visual structure, and feeling the presence of supernatural beings.

One of the foundational tasks of perception is to detect and characterize *structure* in our local environments -- in the form of objects, events, and even abstract statistical regularities. And this can be explored indirectly by asking how people perceive randomness (that is, the lack of structure). Many studies over many decades have explored the ability to detect *whether* a sequence of items is random or not, in a categorical sense. Here, in contrast, we ask how people perceive different *degrees* of randomness, and we provide some counterintuitive answers. People viewed coin-flip outcomes, one by one, and used a slider to report how random the sequence appeared. Different sequences had varying degrees of actual randomness, as determined by bits per symbol, and the probability of an outcome repeat. This allowed us to ask: Are people equally sensitive to all degrees of structure? Or might they be especially sensitive only to whether a sequence has any structure (or no structure at all) -- or to whether a sequence is more or less structured than some average value? In fact, none of these were true. Instead, certain degrees of structure were far more discriminable from each other, even while equating objective differences. For example, when characterized in terms of average information, people were highly sensitive to the difference between 0.6 and 0.7 bits -- while showing no ability to discriminate between other similarly adjacent values. This pattern was surprising even to observers: we collected continuous confidence ratings for each of the randomness judgments, and these diverged from the actual patterns of sensitivity in qualitative ways. In particular, people were confident for extreme degrees of structure, but *not* for extreme degrees of randomness. We conclude that people are sensitive to structure as a continuous variable in nonuniform ways that prioritize some degrees of structure over others.

One of the most impressive and well-known examples of the robustness of perception is biological motion: people reliably see locomoting agents from only a handful of moving dots, in point-light walkers (PLWs). But PLWs are also robust in another, even more impressive way: people reliably see biological motion even when the dot arrangements vary dramatically in size from moment to moment. Like biological motion itself, this has been attributed to special features of *bodies* -- which always maintain the same constrained skeletal structure, and engage in characteristic cyclic motion patterns (e.g. pendular motion when walking). Recently, however, many of the most foundational features of biological motion (e.g. spontaneous perception, resistance to noise, disruption by spatial scrambling, and extensions to rich secondary properties) have also been observed for a very different form of non-rigid motion: point-light *cloths* (PLCs) -- as when a ribbon (or a sheet on a clothesline) is waving in the wind. Are PLCs similarly robust? To find out, we showed people point-light displays of dynamic soft materials that changed radically from moment-to-moment in both size (as in previous PLW research) and location (not previously explored). Despite these constant drastic changes, people still reliably and spontaneously perceived the motions of cloths. We demonstrate this in a series of both psychophysical experiments and phenomenological demonstrations, and we show how this also extends to secondary properties such as cloth stiffness. These results are perhaps even more impressive than with PLWs, since cloths can vary dynamically so much more than bodies (due to the lack of any skeletal structure or cyclical motion). This form of robust perception in point-light displays is not specific to bodies after all, but may rather reflect more general mechanisms of extracting structure from nonrigid motion, perhaps incorporating intuitive physical constraints.

Suppose you go jogging while wearing a ring on your finger, and earphones in your ears. Will the ring fall off, or will the earphones fall out? The answers will depend not so much on the relation itself ('on', 'in'), but on the degree to which the ring/earphones fit *tightly* vs. *loosely*. This distinction is so important that it can even have lexical priority -- e.g. in Korean, where you might use the same term to indicate putting a ring *on* your finger, or earphones *in* your ears, as long as the fit is similarly tight. Given the power of this distinction, might it also be prioritized even in visual processing? To find out, we exploited the logic of *categorial perception*, with simple animations of a cylinder moving down into a container. English speakers viewed sequential pairs of such events , and simply reported whether they were identical, vs. different in any way. Both the cylinder and the container could vary in diameter across animations, to implement tight fits (e.g. narrow cylinders moving into narrow containers) or loose fits (e.g. narrow cylinders moving into wide containers). Observers were much better at noticing Cross-Type changes (e.g. from a tight-fit to a loose-fit animation) than Within-Type changes (e.g. two different tight-fit animations), even though (1) the tight/loose distinction was always task-irrelevant, and (2) Within-Type changes were always greater in objective magnitude. Moreover, these results were specific to the tight vs. loose fits themselves, since these effects disappeared when the same movements (and sizes) occurred in the context of occlusion (where fit is irrelevant) instead of containment (where fit is consequential). This work thus demonstrates a new form of intuitive physics that may serve as an underlying visual primitive for dynamic event perception.

Recently, multivariate decoding of EEG data has identified a signal that scales with the number of items in working memory (WM), regardless of the specific content being maintained or the number of spatially attended locations. One possibility is that this signal reflects an abstract indexing process that binds the content of items to their context in space and time for maintenance and accessibility. To explore this possibility, we examined whether a similar load signal exists for "featureless objects", which can only be described by their coordinates in space and time. On each trial, participants viewed a dense grid of crosses of random orientations. A moving object was implemented by having a single cross change from one random orientation to another, with changes occurring between adjacent crosses across frames. These transients yielded a persisting trackable object, even though (a) there is no constant surface feature across frames, and (b) it is impossible to identify objects on any static frame. In 2 EEG datasets, participants completed a task in which they tracked either 1 or 2 cued featureless objects, and a WM task in which they remembered either colors or the shape and location of "dot cloud" stimuli while spatial attention was controlled for. In both experiments, EEG decoding found a stable signal that scaled with the number of tracked featureless objects. However, we found no consistent evidence that the tracking load signal and the WM load signal generalized to one another. Although planned studies will examine whether this conclusion generalizes to traditional tracking tasks that may rely on other mechanisms (Lu and Sperling, 1996), these results suggest that distinct mechanisms may guide WM storage and attentional tracking.

Visuomotor adaptation is thought to be driven by sensory prediction errors -- the difference between predicted and observed visual feedback given an outgoing motor command. In typical tasks in this domain, visual feedback is given in the form of objects such as a white cursor moving on a black background. As a result, the ensuing adaptation could be driven either by the persisting lower-level visual feature (a white patch moving across a black background), or by the resulting object representation itself. To find out requires a stimulus that at first blush may sound impossible: a persisting object without any persisting features. That is what we explored in the present project, exploiting "featureless objects." In Experiment 1, subjects moved their cursor in a workspace tiled with blue squares of varying brightness. As they moved, each square their hand entered changed from one random shade to another. These brief transients supported the perception of a persistent, trackable object, even though no surface feature was preserved across frames and no static frame allowed for object identification. Feedback was then rotated relative to the angle of subjects' hand movements, and adaptation was measured via implicit aftereffects (i.e., automatic compensatory adjustments to the induced errors). Subjects adapted robustly, and aftereffects were larger under featureless object feedback versus conventional cursor feedback. In Experiment 2, we extended this finding to another form of featureless object feedback: random pixel shuffling within a static background. We again observed adaptation, and observed a canonical cosine-shaped generalization function when aftereffects were measured at nearby reaching directions not experienced during training. In Experiment 3, we discovered that different forms of featureless object feedback could be tied to the same visuomotor memory, observing complete transfer of adaptation between the two feedback contexts. We conclude that implicit visuomotor adaptation is fundamentally an object-based process.

Memory for a visual stimulus depends not only on the observer, but also on the stimulus itself: some objects or events seem to stick in our minds much more readily than others. But just how can -- and should -- this type of visual *memorability* be measured? A tidal wave of recent work has almost always employed either 'corrected recognition' (hit rate minus false-alarm rate) or d' (as a measure of signal-detection sensitivity). Here we show how these measures actually capture a mixture of memorability *and response bias* -- and in the worst cases may degenerate into pure measures of bias. We demonstrate this in three steps. First, we show how this problem can occur in principle, in constructed examples. Second, we show how it operates in practice, in a set of actual case studies from the published empirical literature -- highlighting examples both where this problem is tempered, and where it is catastrophic. Third, we survey the extent of this problem across the literature, in an analysis of 34 previous memorability experiments. We find that the clear majority of experiments show an unexpected positive correlation between hits and false alarms (indicative of response bias), and that inadvertently measuring response bias is ubiquitous. To address the serious limitations of existing measures, we introduce a new way of measuring memorability based on estimating "memory axes". This approach makes explicit the underlying covariance structure of the data, using principal-components analysis to directly orthogonalize memorability from response bias. We suggest that future work on memorability should aim to separate response bias and memorability as distinct dimensions of stimulus-driven memory performance. Doing so may profoundly change the theoretical interpretation of visual memorability studies.