Resolving human object recognition in space and time

Cichy, Radoslaw Martin; Pantazis, Dimitrios; Oliva, Aude

doi:10.1038/nn.3635

Cited by 745 publications

(1,076 citation statements)

References 52 publications

Supporting

122

Mentioning

949

Contrasting

Order By: Relevance

“…We found that the time course rose sharply after image onset, reaching significance at 50 ms (45-52 ms) and a peak at 97 ms (94-102 ms). This indicates that single scene images were discriminated early by visual representations, similar to single images with other visual content (Thorpe et al, 1996;Carlson et al, 2013;Cichy et al, 2014;Isik et al, 2014), suggesting a common source in early visual areas (Cichy et al, 2014).…”

Section: Neural Representations Of Single Scene Images Emerged Earlymentioning

confidence: 84%

“…We characterized the emerging representation of scenes in the human brain using multivariate pattern classification methods (Carlson et al, 2013;Cichy et al, 2014) and representational similarity analysis (Kriegeskorte, 2008;Kriegeskorte and Kievit, 2013) on combined MEG and computational model data. We found that neural representations of individual scenes and the low-level image property contrast emerged early, followed by the scene layout property scene size at around 250 ms.…”

Section: Discussionmentioning

confidence: 99%

“…Using multivariate pattern classification (Carlson et al, 2013;Cichy et al, 2014;Isik et al, 2014) and representational similarity analysis (Kriegeskorte, 2008;Kriegeskorte and Kievit, 2013;Cichy et al, 2014) on millisecond-resolved magnetoencephalography data (MEG), we identified a marker of scene size around 250 ms, preceded by and distinct from an early signal for lower-level visual analysis of scene images at ~100ms. Furthermore, we demonstrated that the scene size marker was independent of both low-level image features (i.e.…”

Section: The Temporal Dynamics Of Spatial Layout Processingmentioning

confidence: 99%

“…Representational clustering analysis for size-Interpreting decoding accuracy as a measure of dissimilarity between patterns, and thus as a distance measure in representational space (Kriegeskorte and Kievit, 2013;Cichy et al, 2014), we partitioned the RDM decoding matrix into within-and between-level segments for the factor scene size (Fig. 2A).…”

Section: Multivariate Pattern Classification Of Meg Datamentioning

confidence: 99%

“…To determine the extent to which visual representations learned by the deep scene model and the human brain are comparable, we used representational similarity analysis (Kriegeskorte, 2008;Cichy et al, 2014). The key idea is that if two images evoke similar responses in the model, they should evoke similar responses in the brain, too.…”

Section: Neural Representations Of Single Scene Images Were Predictedmentioning

confidence: 99%

See 4 more Smart Citations

Dynamics of scene representations in the human brain revealed by magnetoencephalography and deep neural networks

Cichy

Khosla

Pantazis

et al. 2015

Preprint

Self Cite

View full text Add to dashboard Cite

Human scene recognition is a rapid multistep process evolving over time from single scene image to spatial layout processing. We used multivariate pattern analyses on magnetoencephalography (MEG) data to unravel the time course of this cortical process. Following an early signal for lowerlevel visual analysis of single scenes at ~100 ms, we found a marker of real-world scene size, i.e. spatial layout processing, at ~250 ms indexing neural representations robust to changes in unrelated scene properties and viewing conditions. For a quantitative model of how scene size representations may arise in the brain, we compared MEG data to a deep neural network model trained on scene classification. Representations of scene size emerged intrinsically in the model, and resolved emerging neural scene size representation. Together our data provide a first description of an electrophysiological signal for layout processing in humans, and suggest that deep neural networks are a promising framework to investigate how spatial layout representations emerge in the human brain.

show abstract

Section: Neural Representations Of Single Scene Images Emerged Earlymentioning

confidence: 84%

Section: Discussionmentioning

confidence: 99%

Section: The Temporal Dynamics Of Spatial Layout Processingmentioning

confidence: 99%

Section: Multivariate Pattern Classification Of Meg Datamentioning

confidence: 99%

Section: Neural Representations Of Single Scene Images Were Predictedmentioning

confidence: 99%

See 3 more Smart Citations

Dynamics of scene representations in the human brain revealed by magnetoencephalography and deep neural networks

Cichy

Khosla

Pantazis

et al. 2015

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

Visual Object Recognition

Behrmann

Vida

2018

Stevens' Handbook of Experimental Psychology and Cognitive Neuroscience

View full text Add to dashboard Cite

Humans have the remarkable ability to encode and remember thousands of familiar objects in great detail, and yet the psychological and neural mechanisms that contribute to this facility remain elusive. This review considers the recent progress made on this topic, with consideration given to the data gleaned from a host of relevant methodologies. The results of psychophysical and neuroimaging investigations are considered alongside findings that examine object recognition in individuals with deficits in this domain. We also consider the emergence of object recognition skills through behavioral and imaging studies with children. Theoretical accounts and new techniques, including representational (dis)similarity analysis and deep convolution networks, are also considered. As is often the case in such reviews, more questions are raised than answers provided, and the comprehensive understanding of the mechanisms giving rise to visual object recognition remains an alluring challenge to cognitive science.

show abstract

The Advents of Hybrid Imaging Modalities: A New Era in Neuroimaging Applications

et al. 2017

View full text Add to dashboard Cite

Hybrid Imaging modalities have shown great potential in medical imaging and diagnosis. A more comprehensive and targeted view of neurological disorders can be achieved by blending the anatomical and functional perspectives through hybridization. With consistently improving technologies, there have been many developments in fused imaging techniques over the past few decades. This article provides an overview of various bimodal and trimodal hybrid imaging techniques being developed and explored for neuroimaging applications. Recent advancements and potentials are discussed for single photon emission computed tomography‐computed tomography (SPECT‐CT), positron emission tomography‐CT (PET‐CT), PET‐magnetic resonance imaging (PET‐MRI), electroencephalography‐functional magnetic resonance imaging (EEG‐fMRI), magnetoencephalography‐fMRI (MEG‐fMRI), EEG‐near‐infrared spectroscopy (EEG‐NIRS), magnetic resonance‐PET‐EEG (MR‐PET‐EEG) and MR‐PET‐CT in the perspective of neuroimaging. A comparison of these hybrid approaches is provided on a single platform to analyze their performance on the basis of several common factors essential for imaging and analyzing neurological disorders and in vivo molecular processes. This article also provides an overview of recently developed advanced imaging technologies that are being hybridized with other imaging modalities and being explored as potential techniques for neuroscience. Novel approaches and clinical applications of hybrid neuroimaging are anticipated with inclusion of new technologies, better sensing capabilities, multimodal probes, and improved hybridization.

show abstract

Resolving human object recognition in space and time

Cited by 745 publications

References 52 publications

Dynamics of scene representations in the human brain revealed by magnetoencephalography and deep neural networks

Dynamics of scene representations in the human brain revealed by magnetoencephalography and deep neural networks

Visual Object Recognition

The Advents of Hybrid Imaging Modalities: A New Era in Neuroimaging Applications

Contact Info

Product

Resources

About