Despite the very different retinal images that result from different viewing conditions, humans have little difficulty recognising visual objects in varying circumstances. One source of variability is 2-D rotation, which results in an object having different orientations. Here, we studied how the brain transforms rotated object images into object representations that are tolerant to rotation. We measured time-varying electroencephalography responses to object images shown in eight different orientations, presented at either 5 Hz or 20 Hz. We used multivariate classification to assess when rotation-tolerant object information emerged, and whether the rotation-tolerant processing would be limited at the faster presentation rate. We compared this to fixed-rotation measures of object decoding, where the classifier is trained and tested on the same orientation. Our results showed that both fixed-rotation and rotation-tolerant object decoding emerged at an early stage of processing, less than 100 ms after stimulus onset. However, rotation-tolerant information peaked later than fixed-rotation information, suggesting rotation-tolerant object representations are most prominent during a late stage of processing, around 200 ms after stimulus onset. Both fixed-rotation and rotation-tolerant object information was lower for the 20 Hz compared to 5 Hz presentation rate, which suggests that object information processing is disrupted, but not eliminated, for fast presentation rates. Our results show that object information arises at similar times in the brain regardless of whether it is investigated with the fixed-rotation or rotation-tolerant object decoding method. An object representation that is tolerant to rotation and generalises across different exemplars of the same object is established in later stages of processing.

Humans effortlessly relate what they see to what they know, drawing on existing knowledge of the perceptual, conceptual, and contextual attributes of objects while searching for and recognizing objects. Although prior studies have investigated the temporal dynamics of perceptual and conceptual object properties in the neural signal, it remains unclear whether and when contextual associations are uniquely represented. In this study, we used representational similarity analysis on electroencephalography (EEG) data to explore how the brain processes the perceptual, conceptual, and contextual dimensions of object knowledge over time. Using human similarity judgments of 190 naturalistic object concepts presented as either images or words, we constructed separate behavioral models of the perceptual, conceptual, and contextual properties of objects. We correlated these models with neural patterns from two EEG datasets, one publicly available and one newly collected, both recorded while participants passively viewed the same object stimuli. Across both datasets, we found that perceptual features dominated the early EEG response to object images, and conceptual features emerged later. Contextual associations were also reflected in neural patterns, but their explanatory power largely overlapped with that of conceptual models, suggesting limited unique representation of the contextual attributes of objects under passive viewing conditions. These results highlight the integration of perceptual and conceptual information by the brain when processing visual objects. By combining high temporal resolution EEG with behaviorally derived models, this study advances our understanding of how distinct dimensions of object knowledge are encoded in the human brain.

Seeing the world feels effortless, yet it is accomplished by two hemispheres that do not process visual information in the same way. Whether this involves distinct processing cascades, tight hemispheric coupling, or something in between has fundamental implications for understanding the emergence of coherent perception from divided processes. Images presented to the fovea project to both hemispheres, offering an opportunity to isolate intrinsic hemispheric differences in visual coding. Here, we used electroencephalography and neural decoding methods to investigate the dynamics of foveal visual processing in the left and right hemispheres. Human participants (N = 20; 15 females, 5 males) viewed images of objects, faces and words in rapid sequences while performing an orthogonal task. We found different trajectories of visual coding within the left and right hemispheres, and these differences were characterised by distinct featural biases in each hemisphere, with a particular left bias for rectilinearity and right bias for colour. Yet, despite encoding visual information differently, the left and right hemispheres appear to converge on a shared representation. The results provide new insights into hemispheric dynamics underlying visual perception and the complementary roles of the left and right hemispheres in processing visual information.

Visual information plays a key role in guiding food-related decisions. While previous studies have shown that features such as calories and naturalness are encoded by the brain, upon simply seeing the stimuli, it remains unclear how this encoding is shaped by the observer’s current state. In this study, we explore the effect of 1) hunger state, 2) task relevance, and 3) current individual preference on the processing of visual food information. Participants (N = 23) underwent two EEG sessions: one after fasting overnight and another after eating normally. During each session, participants did two separate tasks, one where the stimuli were task-relevant and one where attention was distracted away. We used multivariate analysis methods to assess the impact of hunger on the representation of food-related features, and to determine the time-course of information related to food flavour, personal appeal, and arousal, across both tasks. Results showed that information about edibility (food versus non-food object), food identity (e.g. hamburger versus pizza), flavour profile, or personal appeal and arousal was not influenced by the hunger manipulation. Flavour was represented regardless of attentive state, whereas personal appeal and arousal information emerged later and were only observed when the food was task relevant. We found that food appeal and arousal encoding were more closely aligned with behavioural ratings within rather than between sessions, suggesting the nature of the encoding was driven by current state. The study provides insights into how personal preferences and physiological states influence the representation of food information in the brain.

A mere glimpse of a food can tell us whether it is savoury or sweet, healthy or indulgent, before we have even tasted it. What structure underlies these visual judgements? Are they driven primarily by food categories, such as fruit, grain, or dessert, by food properties that cut across category boundaries, such as sweet versus savoury, or by some combination of the two? We studied this by collecting over 140,000 triplet judgements in which participants indicated which of three foods “doesn’t belong,” enabling us to reconstruct the representational space humans use to differentiate foods. The resulting space revealed two broad axes: a savoury–sweet continuum and a healthy–indulgent continuum. Food category membership was the strongest single organising principle, outperforming any individual flavour or appraisal dimension, while semantic dimensions explained substantial additional structure. A complementary data-driven analysis recovered a similar structure without access to category labels. Together, these findings show that visual food representations are organised around a categorical scaffold, with continuous semantic properties refining similarity within categories and creating links across category boundaries.

The human brain continuously integrates information across its two hemispheres to construct a coherent representation of the perceptual world. Characterizing how visual information is represented in each hemisphere over time is crucial for understanding how hemispheric transfer contributes to perception. Here, we investigated information processing within each hemisphere over time and the degree to which it is distinct or duplicated across hemispheres. We presented participants with object images lateralized to the left or right visual fields while measuring their brain activity with electroencephalography. Stimulus coding was more robust and emerged earlier in the contralateral than the ipsilateral hemisphere. Presentation of two stimuli, one to each hemifield, reduced the fidelity of representations in both hemispheres relative to one stimulus alone, signifying hemispheric interference. Last, we found that processing within the contralateral, but not ipsilateral, hemisphere was biased to image-related over concept-related information. Together, these results suggest that hemispheric transfer operates to filter irrelevant information and efficiently prioritize processing of meaning.

Evidence suggests that mental imagery and veridical perception recruit similar components of the human visual system. If so, neural representations of imagined and real stimuli should interact with one another, combining constructively or competing antagonistically. To determine if and how real and imagined visual stimuli interact in the brain, we asked participants to mentally visualise white bars at specific orientations after a rhythmic countdown while their brain activity was recorded using electroencephalography. Stimuli were imagined in isolation, or while another stimulus at a highly or poorly congruent orientation appeared on-screen. Multivariate pattern analysis was used to assess whether overlap between imagined and real stimulus features enhanced or diminished stimulus-specific sensory information in the brain. Findings showed that imagined and real orientation could be decoded from brain activity, with real orientation decoding mildly amplified by highly congruent, but not poorly congruent, imagined orientations. Although interactions between real and imagined stimuli were observed, no evidence was detected to suggest that imagined and real stimuli use the same neural activity patterns to encode sensory information. Instead, congruent imagery seemed only to amplify activity which had already been induced by real percepts, targeting late- but not early-stage perceptual representations. Ultimately, this study suggests that imagined and real stimuli interact in a mildly constructive manner, with imagination mostly acting in a modulatory capacity.### Competing Interest StatementThe authors have declared no competing interest.

Food is fundamental to survival, and our brains are highly attuned to rapidly process food stimuli. Neural signals show that foods can be discriminated as edible or inedible as early as 85 ms after stimulus onset,1 distinguished as processed or unprocessed beginning at 130 ms,2 and as high or low density from 165 ms.3 Recent evidence revealed specialized processing of food stimuli in the ventral visual pathway,4,5,6 an area that underlies perception of faces and other important objects. For many visual objects, perception can be biased toward recent perceptual history (known as serial dependence7,8). We examined serial dependence for food in two large samples (n > 300) who rated sequences of food images for either "appeal" or "calories." Ratings for calories were highly correlated between participants and were similar for males and females. Appeal ratings varied considerably between participants, consistent with the idiosyncratic nature of food preferences, and tended to be higher for males than females. High-calorie ratings were associated with high appeal, especially in males. Importantly, response biases showed clear positive serial dependences: higher stimulus values in the previous trials led to positive biases, and vice versa. The effects were similar for males and females and for calories and appeal ratings and were remarkably consistent across participants. These findings square with recently found food selectively in the visual temporal cortex, reveal a new mechanism influencing food decision-making, and suggest a new sensory-level component that could complement cognitive strategies in diet intervention.

Little is known about the perceptual characteristics of mental images nor how they vary across sensory modalities. We conducted an exhaustive survey into how mental images are experienced across modalities, mainly targeting visual and auditory imagery of a single stimulus, the letter "O", to facilitate direct comparisons. We investigated temporal properties of mental images (e.g. onset latency, duration), spatial properties (e.g. apparent location), effort (e.g. ease, spontaneity, control), movement requirements (e.g. eye movements), real-imagined interactions (e.g. inner speech while reading), beliefs about imagery norms and terminologies, as well as respondent confidence. Participants also reported on the five traditional senses and their prominence during thinking, imagining, and dreaming. Overall, visual and auditory experiences dominated mental events, although auditory mental images were superior to visual mental images on almost every metric tested except regarding spatial properties. Our findings suggest that modality-specific differences in mental imagery may parallel those of other sensory neural processes.

Children often mistake things that appear to move without visible external intervention, such as clouds, for being alive. Interestingly, these mistakes are still made by adults under time pressure. Goldberg and Thompson-Schill (2009) demonstrated this using a rapid classification task, and showed that adults make more errors when judging if plants and moving natural things are alive (compared to animals, and still or artificial things). Over four experiments, we explored Goldberg and Thompson-Schill’s (2009) proposal that connections formed in childhood between movement and aliveness are used in adulthood to make decisions about aliveness. Our findings replicated Goldberg and Thompson-Schill’s (2009) results, that plants and moving natural things were more difficult to classify by aliveness. In addition, we showed that plants and natural things used in their study were also less familiar and harder to visualise than other categories. Interestingly, participants intuitively believed that moving natural things can move without external intervention, suggesting that associations between movement and aliveness influence decisions made without time pressure. These results suggest that movement is an important part of our conceptual understanding of aliveness, and how we encode objects in our environment.

The basic computations performed in the human early visual cortex are the foundation for visual perception. While we know a lot about these computations, a key missing piece is how the coding of visual features relates to our perception of the environment. To investigate visual feature coding, interactions, and their relationship to human perception, we investigated neural responses and perceptual similarity judgements to a large set of visual stimuli that varied parametrically along four feature dimensions. We measured neural responses using electroencephalography (N = 16) to 256 grating stimuli that varied in orientation, spatial frequency, contrast, and colour. We then mapped the response profiles of the neural coding of each visual feature and their interactions, and related these to independently obtained behavioural judgements of stimulus similarity. The results confirmed fundamental principles of feature coding in the visual system, such that all four features were processed simultaneously but differed in their dynamics, and there was distinctive conjunction coding for different combinations of features in the neural responses. Importantly, modelling of the behaviour revealed that every stimulus feature contributed to perceptual judgements, despite the untargeted nature of the behavioural task. Further, the relationship between neural coding and behaviour was evident from initial processing stages, signifying that the fundamental features, not just their interactions, contribute to perception. This study highlights the importance of understanding how feature coding progresses through the visual hierarchy and the relationship between different stages of processing and perception.

Humans make decisions about food every day. The visual system provides important information that forms a basis for these food decisions. Although previous research has focused on visual object and category representations in the brain, it is still unclear how visually presented food is encoded by the brain. Here, we investigate the time-course of food representations in the brain. We used time-resolved multivariate analyses of electroencephalography (EEG) data, obtained from human participants (both sexes), to determine which food features are represented in the brain and whether focused attention is needed for this. We recorded EEG while participants engaged in two different tasks. In one task, the stimuli were task relevant, whereas in the other task, the stimuli were not task relevant. Our findings indicate that the brain can differentiate between food and nonfood items from ∼112 ms after the stimulus onset. The neural signal at later latencies contained information about food naturalness, how much the food was transformed, as well as the perceived caloric content. This information was present regardless of the task. Information about whether food is immediately ready to eat, however, was only present when the food was task relevant and presented at a slow presentation rate. Furthermore, the recorded brain activity correlated with the behavioral responses in an odd-item-out task. The fast representation of these food features, along with the finding that this information is used to guide food categorization decision-making, suggests that these features are important dimensions along which the representation of foods is organized.

Mental imagery is a process by which thoughts become experienced with sensory characteristics. Yet, it is not clear why mental images appear diminished compared to veridical images, nor how mental images are phenomenologically distinct from hallucinations, another type of non-veridical sensory experience. Current evidence suggests that imagination and veridical perception share neural resources. If so, we argue that considering how neural representations of externally generated stimuli (i.e. sensory input) and internally generated stimuli (i.e. thoughts) might interfere with one another can sufficiently differentiate between veridical, imaginary, and hallucinatory perception. We here use a simple computational model of a serially connected, hierarchical network with bidirectional information flow to emulate the primate visual system. We show that modelling even first approximations of neural competition can more coherently explain imagery phenomenology than non-competitive models. Our simulations predict that, without competing sensory input, imagined stimuli should ubiquitously dominate hierarchical representations. However, with competition, imagination should dominate high-level representations but largely fail to outcompete sensory inputs at lower processing levels. To interpret our findings, we assume that low-level stimulus information (e.g. in early visual cortices) contributes most to the sensory aspects of perceptual experience, while high-level stimulus information (e.g. towards temporal regions) contributes most to its abstract aspects. Our findings therefore suggest that ongoing bottom-up inputs during waking life may prevent imagination from overriding veridical sensory experience. In contrast, internally generated stimuli may be hallucinated when sensory input is dampened or eradicated. Our approach can explain individual differences in imagery, along with aspects of daydreaming, hallucinations, and non-visual mental imagery.

The human brain integrates information across the hemispheres to construct a coherent representation of the world. Characterising how visual information is coded in each hemisphere is crucial for understanding the nature of information transfer in the brain. Here, we investigated information processing within each hemisphere and its distinctiveness across hemispheres. We presented participants (N = 20) with images of faces, words and objects in rapid sequences while neural responses were measured using electroencephalography (EEG). To drive distinct responses in each hemisphere, stimuli were presented either centrally or lateralised to the left or right visual fields. Participants performed an orthogonal colour change task on dots that marked possible image positions. Multivariate pattern analyses were applied to the neural data to assess coding of object information in the brain, separately for electrode clusters over each hemisphere. Stimulus information was more robust and emerged earlier in the contralateral than the ipsilateral hemisphere. Interestingly, the temporal dynamics within the two hemispheres followed different trajectories. Representational structure was consistent across the hemispheres with delays approximating interhemispheric transmission time. These results provide insights into the dynamics of object perception and the competitive versus cooperative nature of hemispheric processing.

Many studies have shown that recent perceptual history can bias present perception, known as serial dependence. We examined this phenomenon for sequences of food images that participants rated for either ‘appeal’ or ‘calories’ and found strong positive dependencies. A large sample in an online experiment (n>300) made 450 ratings of food images (150 stimuli, randomly presented three times), rating appeal and calories in separate counterbalanced blocks. A first analysis based on the previous trial’s rating showed that for both appeal and calories there were clear assimilations towards the preceding trial’s rating which were well described by a difference of Gaussian model. The amplitude of the serial effect for appeal was roughly twice that for calories. For food appeal, the serial effect was greater in males than females. There were no sex differences for calorie ratings. These serial analyses were based on previous response, which emphasises perceptual decision. A second analysis based on the previous stimulus (using the mean of all ratings of that image) was conducted. Only the serial effect for calories was significant on the stimulus analysis, possibly indicating a perceptual effect of caloric value of food. Various demographic data and scales were recorded before the experiment and two correlations survived multiple comparisons. For appeal ratings, males showed greater amplitude of serial effect with increasing tiredness. For calorie ratings, all subjects showed greater serial bandwidth with increasing BMI.

Patterns of brain activity contain meaningful information about the perceived world. Recent decades have welcomed a new era in neural analyses, with computational techniques from machine learning applied to neural data to decode information represented in the brain. In this article, we review how decoding approaches have advanced our understanding of visual representations and discuss efforts to characterize both the complexity and the behavioral relevance of these representations. We outline the current consensus regarding the spatiotemporal structure of visual representations and review recent findings that suggest that visual representations are at once robust to perturbations, yet sensitive to different mental states. Beyond representations of the physical world, recent decoding work has shone a light on how the brain instantiates internally generated states, for example, during imagery and prediction. Going forward, decoding has remarkable potential to assess the functional relevance of visual representations for human behavior, reveal how representations change across development and during aging, and uncover their presentation in various mental disorders.

Although mental imagery is often studied as a visual phenomenon, it can occur in any sensory modality. Given that mental images may recruit similar modality-specific neural systems to those which support veridical perception, the properties of mental images may be constrained by the modality in which they are experienced. Yet, little is known about how mental images are experienced at all, let alone how such experiences may vary depending on the modality in which they occur. Here we explored how mental images are experienced in different modalities using an extensive questionnaire. Mainly focusing on visual and auditory mental imagery, we surveyed participants on if and how they experienced their thought content in a sensory way when thinking about the appearance or sound of the letter “O”. Specifically, we investigated temporal properties of imagined content (e.g. onset latency, duration), as well as spatial properties (e.g. apparent location), effort (e.g. ease, spontaneity, control), dependence on body movements (e.g. eye movements), interactions between real and imagined content (e.g. inner speech during reading), the perceived normality of imagery experiences, and how participants labeled their own experiences. Participants also ranked their mental imagery experiences in the five traditional sensory modalities and reported on the involvement of each modality during their thoughts, imagination, and dreams. Confidence ratings were taken for every answer recorded. Overall, visual and auditory experiences tended to dominate mental events relative to other sensory modalities. However, most people reported that auditory mental imagery was superior to visual mental imagery on almost every metric tested, except with respect to spatial properties. Our findings suggest that mental images are restrained in a similar matter to other modality-specific sensory processes in the brain. Broadly, our work also provides a wealth of insights and observations into how mental images are experienced by individuals, acting as a useful resource for future investigations.

Selective attention prioritises relevant information amongst competing sensory input. Time-resolved electrophysiological studies have shown stronger representation of attended compared to unattended stimuli, which has been interpreted as an effect of attention on information coding. However, because attention is often manipulated by making only the attended stimulus a target to be remembered and/or responded to, many reported attention effects have been confounded with target-related processes such as visual short-term memory or decision-making. In addition, attention effects could be influenced by temporal expectation about when something is likely to happen. The aim of this study was to investigate the dynamic effect of attention on visual processing using multivariate pattern analysis of electroencephalography (EEG) data, while (1) controlling for target-related confounds, and (2) directly investigating the influence of temporal expectation. Participants viewed rapid sequences of overlaid oriented grating pairs while detecting a "target" grating of a particular orientation. We manipulated attention, one grating was attended and the other ignored (cued by colour), and temporal expectation, with stimulus onset timing either predictable or not. We controlled for target-related processing confounds by only analysing non-target trials. Both attended and ignored gratings were initially coded equally in the pattern of responses across EEG sensors. An effect of attention, with preferential coding of the attended stimulus, emerged approximately 230 ms after stimulus onset. This attention effect occurred even when controlling for target-related processing confounds, and regardless of stimulus onset expectation. These results provide insight into the effect of feature-based attention on the dynamic processing of competing visual information.