Abstract Volitional intention can bias perception in cases where two or more interpretations of a stimulus are available to us. The neural mechanisms whereby such an intention influences perception are poorly understood. Here we investigated whether intending to see horizontal versus vertical motion in a subsequently presented instantaneous position shift of a quartet apparent motion stimulus establishes decodable sensory representations prior to both the position shift and the perception of motion. Twelve participants underwent fMRI scanning under three conditions: (1) while passively viewing either continuously or (2) discretely moving quartet stimuli, or (3) while actively intending to see a subsequent single-shot apparent motion as either a vertical or horizontal motion. Multivariate decoding analyses revealed that activity patterns during the intention period of (3) generalized to patterns evoked by both (1) physical and (2) ambiguous motion perception. Cross-decoding was strongest within dorsal/lateral visual regions, including hMT+, V3AB, and the intraparietal sulcus (IPS), but was largely absent from ventral visual cortex. Widespread overlap was also observed between intention-related and perceptual motion representations throughout the dorsal/lateral visual cortex. Our findings suggest that volitional intention establishes prospective sensory representations before perceptual experience emerges and that these representations closely resemble those associated with illusory motion perception. The predominance of intention-related representations within dorsal/lateral visual regions is consistent with top-down influences from attentional control systems. More broadly, the results demonstrate that internally generated cognitive states can shape sensory representations, constraining subsequent perceptual experience. Significance Statement How thoughts and intentions influence perception is an important question in the cognitive neuroscience of consciousness. By combining fMRI with multivariate decoding, we demonstrate that volitional intention to see a subsequent instantaneous position shift as either horizontal or vertical motion recruits sensory representations that resemble those evoked during passive perception, with the strongest effects occurring in dorsal visual and parietal cortex. These findings indicate that top-down signals can proactively configure sensory representations and potentially bias the perceived direction of apparent motion, providing new insight into the neural mechanisms through which intention influences conscious visual experience.

Mirror-mediated localization of hidden objects is well documented in vertebrates1,2,3,4,5,6,7,8,9,10,11,12 but has never been demonstrated in invertebrates. Using mirrors to locate otherwise occluded objects is a form of mediated perception, linking a visible reflection to an occluded location13,14 and is seen by some as a precursor to self-recognition.15 Cephalopods offer a fascinating test case of convergent cognition, having independently evolved sophisticated perceptual and cognitive abilities that are similar to mammals,16,17,18 after diverging from a common ancestor over 520 million years ago.19 In addition, they react to mirror images as though they were conspecifics.20,21,22,23 We projected a virtual crab that was visible only via mirror reflection onto a tank wall. Three Octopus bimaculoides were trained to navigate to the projection site instead of the mirror. All three octopuses learned this task, successfully choosing the correct side in 73% of trials. Critically, octopuses sometimes moved away from the visible reflection and climbed over the side walls of the start chamber to reach visually occluded locations that were spatially aligned with the reflected prey location. This behavior suggests (1) the ability to inhibit a direct approach to salient visual stimuli, and (2) a spatial representation that integrates mirror information with knowledge of 3D tank geometry. These findings extend mirror-use capabilities to invertebrates, demonstrating that cephalopods can employ mirror reflections for spatial navigation. The independent evolution of cognitive capacities underlying mirror use across diverse taxa suggests that common solutions may have evolved to solve spatial navigation challenges.

The difficulty of tracking multiple moving objects among identical distractors increases with the number of tracked targets. Previous research has shown that the number of targets tracked (i.e., load) modulates activity in brain areas related to visuospatial attention, giving rise to so-called attention response functions (ARFs). While the hemifield/hemispheric effects of spatial attention (e.g., hemispatial neglect, hemifield capacity limits) are well described, it had not previously been tested whether a hemispheric or hemifield imbalance exists among ARFs. By recording blood oxygenation level-dependent activity from human brains (n =19, female and male) in a multiple-object tracking paradigm, we show that the number of tracked objects modulates activity in a large network of areas bilaterally. A significant effect of contralateral load was found in earlier areas throughout the dorsal and ventral visual streams, while the effects of ipsilateral load emerged in later areas. Both contra-and ipsilateral load significantly influenced activity in the parietal and frontal lobes, specifically the dorsal attention network. In addition, some brain regions in the occipital lobe were significantly more sensitive to contralateral than ipsilateral load. Our results are consistent with findings showing that a diverse set of brain areas contributes to tracking multiple targets. In particular, we extend the canonical view of load-based ARFs to include hemifield bias. Given the hemifield-specific nature of speed and capacity limits to multiple-object tracking, we conjecture that areas that show a strong hemifield preference may impose a bottleneck on processing that results in limits on the capacity and speed of tracking.

A central debated question in the study of object-based attention (OBA) is whether attention to the object-mediated deployment of attention is obligatory and automatic [Chen, Z., & Cave, K. R. Reinstating object-based attention under positional certainty: The importance of subjective parsing. Perception & Psychophysics, 68, 992-1003, 2006] or whether the pattern of results is driven by other non-obligatory factors, such as prioritization of invalid target locations [Shomstein, S., & Yantis, S. Object-based attention: Sensory modulation or priority setting? Perception & Psychophysics, 64, 41-51, 2002]. However, virtually all behavioral measures attributed to OBA are based on examining performance on invalid-cue trials, the inclusion of which confounds the assessment of the automaticity hypothesis. Our approach to resolve this issue is to determine whether effects of OBA can be observed in a 100% valid cueing paradigm. In this article, we investigate the obligatory nature of OBA by leveraging the spatial specificity of fMRI and the retinotopic organization of early visual cortex. We aimed to identify potential neural correlates of OBA in the complete absence of invalid trials. Participants perform a version of the classic two-rectangle OBA paradigm while we simultaneously measure changes in BOLD signals arising from retinotopically organized cortical areas V1, V2, and V3. In the first half of the experiment, we used the classic two-rectangle OBA paradigm except that the cue was 100% valid. In the second half, we reduced cue validity to more closely match standard OBA paradigms (runs containing invalid trials). We analyzed BOLD signals arising from our ROIs in V1, V2, and V3 according to their topographic correspondences with the ends of the rectangles in the visual field and compared these. We then compared responses in each ROI according to where the cue had occurred (cued, uncued-same-object, uncued-other-object location). We replicated this procedure in Experiment 2, but changed the layout of the two rectangles from a vertical to a horizontal configuration. Critical result: We observed statistically significant effects of OBA in V3 (Experiment 1) and V1-2 (Experiment 2) in both the 100% valid runs and in runs containing invalid trials. Moreover, the effects of OBA were no smaller in the 100% runs compared with runs containing invalid trials. Conclusion: We see BOLD modulation at the uncued locations consistent with neural correlates of OBA.

Our premodern ancestors had perceptual, motoric, and cognitive functional domains that were modularly encapsulated. Some of these came to interact through a new type of cross-modular binding in our species. This allowed previously domain-dedicated, encapsulated motoric and sensory operators to operate on operands for which they had not evolved. Such operators could at times operate nonvolitionally, while at other times they could be governed volitionally. In particular, motoric operations that derive from the same circuits that compute hand motions for object manipulation could now be retooled for virtual manipulation in a mental workspace in the absence of any physical hand or other effector movements. I hypothesize that the creativity of human imagination and mental models is rooted in premotor simulation of sequential manipulations of objects and symbols in the mental workspace, in analogy with the premotor theory of attention, which argues that attention evolved from "internalized" eye movement circuitry. Overall, operator "disencapsulation" led to a bifurcation of consciousness in humans: a concrete form centered on perception of the body in the physical world and an abstract form focused on explanatory mental models. One of the consequences of these new abilities was the advent of psychotic disorders that do not exist in species possessed solely of the concrete type of consciousness.

Epistemological and ontological conceptions of information are contrasted. The former are based on acts of decoding of extrinsic inputs that result in a decoder becoming informed. The latter are based on intrinsic states or state changes of the system independent of any external factors such as inputs to the system. Ontological conceptions of information, such as those that underlie integrated information theory or any theory that allies itself with panpsychism, are not able to account for consciousness. In the only physical systems that are known to be conscious, namely, animal brains, acts of decoding extrinsic inputs are central to creating consciousness and its contents. Moreover, only a very specific subset of decodings should realize consciousness, because consciousness in animals evolved to create an evaluative experience of what is intrinsically true about the world and the body, which is then used in a perception-action cycle that affords choices among options for behaving in the world in order to accomplish goals.

Learning and recognition can be improved by sorting novel items into categories and subcategories. Such hierarchical categorization is easy when it can be performed according to learned rules (e.g., "if car, then automatic or stick shift" or "if boat, then motor or sail"). Here, we present results showing that human participants acquire categorization rules for new visual hierarchies rapidly, and that, as they do, corresponding hierarchical representations of the categorized stimuli emerge in patterns of neural activation in the dorsal striatum and in posterior frontal and parietal cortex. Participants learned to categorize novel visual objects into a hierarchy with superordinate and subordinate levels based on the objects' shape features, without having been told the categorization rules for doing so. On each trial, participants were asked to report the category and subcategory of the object, after which they received feedback about the correctness of their categorization responses. Participants trained over the course of a one-hour-long session while their brain activation was measured using functional magnetic resonance imaging. Over the course of training, significant hierarchy learning took place as participants discovered the nested categorization rules, as evidenced by the occurrence of a learning trial, after which performance suddenly increased. This learning was associated with increased representational strength of the newly acquired hierarchical rules in a corticostriatal network including the posterior frontal and parietal cortex and the dorsal striatum. We also found evidence suggesting that reinforcement learning in the dorsal striatum contributed to hierarchical rule learning.

A novel haptic illusion is described where deformations of the fingertip skin lead to subsequent misperceptions of an object's shape.

The identification of animal behavior in video is a critical but time-consuming task in many areas of research. Here, we introduce DeepAction, a deep learning-based toolbox for automatically annotating animal behavior in video. Our approach uses features extracted from raw video frames by a pretrained convolutional neural network to train a recurrent neural network classifier. We evaluate the classifier on two benchmark rodent datasets and one octopus dataset. We show that it achieves high accuracy, requires little training data, and surpasses both human agreement and most comparable existing methods. We also create a confidence score for classifier output, and show that our method provides an accurate estimate of classifier performance and reduces the time required by human annotators to review and correct automatically-produced annotations. We release our system and accompanying annotation interface as an open-source MATLAB toolbox.

Selective attention has a capacity limit and this can be reached during multiple object tracking by, among other things, increasing the number of tracked targets. Interestingly, this effect is hemifield specific (Alvarez & Cavanagh, 2005). When targets are distributed across the visual field, more can be tracked than when these same targets are confined to one hemifield. We used fMRI to investigate whether this effect is reflected in the brain activity associated with multiple object tracking. Increases in BOLD activity that are caused by an increase in the number of tracking targets are known as Attention Response Functions (Culham et al., 2001; Jovicich et al., 2001). This monotonic increase of activity with load is well-known and well-replicated, especially in the parietal lobe. Does this effect also reflect the hemifield independence that is seen behaviorally? Participants tracked between zero and two targets among identical looking distractors separately in each hemifield, leading to a total of between zero and four targets confined to their respective hemifields. This allowed us to manipulate the tracking load in each brain hemisphere separately. If the hemispheres are fully independent, then varying load in one hemifield should not impact the BOLD response to the other. We replicate previous findings that BOLD activity increases with load in the intraparietal sulcus and the superior parietal lobule. With regards to hemisphere independence, the results showed considerable heterogeneity. We discuss our findings with respect to the regions of interest that have been identified for attention response functions in previous literature.

The allocation of attention to objects raises several intriguing questions: What are objects, how does attention access them, what anatomical regions are involved? Here, we review recent progress in the field to determine the mechanisms underlying object-based attention. First, findings from unconscious priming and cueing suggest that the preattentive targets of object-based attention can be fully developed object representations that have reached the level of identity. Next, the control of object-based attention appears to come from ventral visual areas specialized in object analysis that project downward to early visual areas. How feedback from object areas can accurately target the object’s specific locations and features is unknown but recent work in autoencoding has made this plausible. Finally, we suggest that the three classic modes of attention may not be as independent as is commonly considered, and instead could all rely on object-based attention. Specifically, studies show that attention can be allocated to the separated members of a group—without affecting the space between them—matching the defining property of feature-based attention. At the same time, object-based attention directed to a single small item has the properties of space-based attention. We outline the architecture of object-based attention, the novel predictions it brings, and discuss how it works in parallel with other attention pathways.

Stroboscopic (strobe) hallucinations are geometric percepts elicited by flickering full-field illumination at specific frequencies. They bear a striking resemblance to the kaleidoscopic imagery induced by hallucinogenic psychedelics. As such, they have the potential to serve as an experimental model for studying these altered states of consciousness in a controlled manner. However, it remains unclear if these similar perceptual experiences share the same underlying neural correlates. To address this question, we studied three EEG signatures associated with hallucinogen administration: alpha power reduction, increased signal complexity, and the flattening of the 1⁄f slope. These EEG signatures were measured during hallucinatory (11Hz - 19Hz) and non-hallucinatory (51Hz – 59Hz) strobe stimulation. It was hypothesized that hallucinatory stimulation would elicit EEG modulations like those observed during hallucinogen administration. Our findings revealed that strobe stimulation induced a brain state remarkably similar to the one induced by hallucinogens. This brain state persists even after strobe stimulation has ceased. However, this brain state was elicited by both hallucination and non-hallucination inducing frequencies. Our results suggest that stroboscopic stimulation, in general, offers a viable experimental model for studying the hallucinogenic brain. However, new measures are needed to serve as neural correlates of geometric hallucinations.

When two pre-existing, separated squares are connected by the sudden onset of a bar between them, viewers do not perceive the bar to appear all at once. Instead, they see an illusory morphing of the original squares over time. The direction of this transformational apparent motion (TAM) can be influenced by endogenous attention deployed before the appearance of the connecting bar. Here, we investigated whether the influence of endogenous attention on TAM results from operations over high-level feature-independent shape representations, or instead over lower level shape representations defined by specific visual features. To do so, we tested the influence of endogenous attention on TAM in first- and second-order displays, which shared common shapes but had different shape-defining attributes (luminance and texture contrast, respectively). In terms of both the magnitude of directional bias and timing, we found that endogenous attention exerted a similar influence on both first- and second-order objects. These results imply that endogenous attention biases the perceived direction of TAM by operating on high-level shape representations that are invariant to the low-level visual features that define them. Our results support a four-stage model of TAM, where a feature encoding stage passes a features-specific layout to a parsing stage that forms discrete, high-level meta-featural shapes, which are then matched and visually interpolated over time.

The apparent motion quartet is an ambiguous stimulus where motion is typically seen either vertically or horizontally. Previous studies have claimed that only one direction can be seen at a time. Here we report that it is possible to perceive both vertical and horizontal motion simultaneously (i.e., a mixed percept). In Experiment 1, participants passively viewed a single presentation of two frames of the quartet motion sequence, then reported whether they experienced vertical motion, horizontal motion, or “both”. We identified the aspect ratio where each participant was equally likely to report vertical or horizontal motion – the point of subjective equality (PSE). Across all aspect ratios, participants reported “both” on an average of 11% of trials. The mean aspect ratio across all “both” trials did not differ significantly from the PSE. Experiment 2 examined volitional perception, the ability to will to see one or the other direction. When the stimulus was set to their PSE from Experiment 1, participants had 75% success in willing horizontal motion (better than chance) but failed to will vertical motion better than chance rates. Interestingly, these vertical trials produced more “both” percepts, though fewer than in passive viewing (5.4%). Our results show that it is possible to simultaneously see horizontal and vertical motion in apparent motion quartets as a split in both directions away from the two initial corners. During passive viewing, these split percepts were most common near the PSE. During volitional perception, the split percepts were reduced in frequency and more frequent when willing vertical than horizontal motion.

A pioneering study by Volkmann (1858) revealed that training on a tactile discrimination task improved task performance, indicative of tactile learning, and that such tactile learning transferred from trained to untrained body parts. However, the neural mechanisms underlying tactile learning and transfer of tactile learning have remained unclear. We trained groups of human subjects (female and male) in daily sessions on a tactile discrimination task either by stimulating the palm of the right hand or the sole of the right foot. Task performance before training was similar between the palm and sole. Posttraining transfer of tactile learning was greater from the trained right sole to the untrained right palm than from the trained right palm to the untrained right sole. Functional magnetic resonance imaging (fMRI) and multivariate pattern classification analysis revealed that the somatotopic representation of the right palm in contralateral primary somatosensory cortex (SI) was coactivated during tactile stimulation of the right sole. More pronounced coactivation in the cortical representation of the right palm was associated with lower tactile performance for tactile stimulation of the right sole and more pronounced subsequent transfer of tactile learning from the trained right sole to the untrained right palm. In contrast, coactivation of the cortical sole representation during tactile stimulation of the palm was less pronounced and no association with tactile performance and subsequent transfer of tactile learning was found. These results indicate that tactile learning may transfer to untrained body parts that are coactivated to support tactile learning with the trained body part. SIGNIFICANCE STATEMENT Perceptual skills such as the discrimination of tactile cues can improve by means of training, indicative of perceptual learning and sensory plasticity. However, it has remained unclear whether and if so, how such perceptual learning can occur if the training task is very difficult. Here, we show for tactile perceptual learning that the representation of the palm of the hand in primary somatosensory cortex (SI) is coactivated to support learning of a difficult tactile discrimination task with tactile stimulation of the sole of the foot. Such cortical coactivation of an untrained body part to support tactile learning with a trained body part might be critically involved in the subsequent transfer of tactile learning between the trained and untrained body parts.

There appear to be three independent systems for allocating attention: space-based, feature based, and object-based. Here, we review the literature of object-based attention to determine its underlying mechanisms. First, findings from unconscious priming and cuing suggest that the pre-attentive targets of object-based attention can be fully developed object representations. Next, the control of object-based attention appears to come from ventral visual areas specialized in object analysis that project downward to early visual areas. Whether feedback from object areas can accurately target the object’s specific locations and features is controversial, but recent work in autoencoding has made this plausible. Finally, we suggest that the three classic modes of attention may not be as independent as is commonly considered, and instead could rely on object-based attention for all three modes of selection. Specifically, studies show that attention can spread over the separated members of a group – without affecting the space between them — matching the defining property of feature-based attention. At the same time, object-based attention directed to a single small item has the properties of space-based attention. Nevertheless, the evidence for a parallel, space-based selection controlled through saccade centers is also convincing. We outline the architecture for this combined system and discuss how it works in parallel with other attention pathways.