Invariant visual representation by single neurons in the human brain
By R. Quian Quiroga, Leila Reddy, Gabriel Kreiman, Christof Koch, Itzhak Fried. 2005
1,939 citations recorded by the source corpus.
It takes a fraction of a second to recognize a person or an object even when seen under strikingly different conditions. How such a robust, high-level representation is achieved by neurons in the human brain is still unclear. In monkeys, neurons in the upper stages of the ventral visual pathway respond to complex images such as faces and objects and show some degree of invariance to metric properties such as the stimulus size, position and viewing angle. We have previously shown that neurons in the human medial temporal lobe (MTL) fire selectively to images of faces, animals, objects or scenes. Here we report on a remarkable subset of MTL neurons that are selectively activated by strikingly different pictures of given individuals, landmarks or objects and in some cases even by letter strings with their names. These results suggest an invariant, sparse and explicit code, which might be important in the transformation of complex visual percepts into long-term and more abstract memories.
Subfields: Cognitive Neuroscience, Neuroscience
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Cognitive Neuroscience
- Nikolaus Kriegeskorte
- Xiao-Jing Wang
- James J. DiCarlo
- György Buzsáki
- J. Urgen Schmidhuber
- Wulfram Gerstner
- Stefano Fusi
- Yoshua Bengio
In Neuroscience
- Nikolaus Kriegeskorte
- J. Urgen Schmidhuber
- Jörn Diedrichsen
- György Buzsáki
- Andrew Y. Ng
- James J. DiCarlo
- Wulfram Gerstner
- Christof Koch
Citations in the corpus
- Recurrent World Models Facilitate Policy Evolution (2018, background): “For instance, we are able to observe a scene and remember an abstract description thereof [7, 67].”
- Building high-level features using large scale unsupervised learning (2011, background): “…in the brain is an area of active investigation, but current experimental evidence suggests the possibility that some neurons in the temporal cortex are highly selective for object categories such as faces or hands (Desimone et al., 1984), and perhaps even specific people (Quiroga et al., 2005).”
- Multimodal datasets: misogyny, pornography, and malignant stereotypes (2021, methodology): “…in the CLIP-RN50_4x model was akin to a multimodal Spiderman "neuron" that responded to not just photos of Spiderman in costume and spiders but also sketches of Spiderman and the text “spider” (Thereby drawing parallels with the so-termed Halle Berry / Jennifer Aniston neuron discovery in [74] ).”
- Synthesizing the preferred inputs for neurons in neural networks via deep generator networks (2016, background): “Such neurons are multifaceted: for example, the “Halle Berry neuron” responds to very different stimuli related to the actress—from pictures of her face, to pictures of her in costume, to the word “Halle Berry” printed as text [1].”
- Network Dissection: Quantifying Interpretability of Deep Visual Representations (2017, background): “This three-step process of network dissection is reminiscent of the procedures used by neuroscientists to understand similar representation questions in biological neurons [23].”
- World Models (2018, methodology): “We are able to observe a scene and remember an abstract description thereof (Cheang & Tsao, 2017; Quiroga et al., 2005).”
- Graying the black box: Understanding DQNs (2016, methodology): “, 2014), others try to explain the function computed by each individual neuron (similar to the ”Jennifer Aniston Neuron” (Quiroga et al., 2005)).”
- Multifaceted Feature Visualization: Uncovering the Different Types of Features Learned By Each Neuron in Deep Neural Networks (2016, background): “Neurons in human brains are similarly multifaceted: a “Halle Berry” neuron was found that responds to very different stimuli related to the actress, from pictures of her in costume to her name printed as text (Quiroga et al., 2005).”
- Open-Ended Learning Leads to Generally Capable Agents (2021, background): “, 2019)), • conduct simple concept decoding tests (often done with linear classifier probes (Alain and Bengio, 2017) or single neuron activity analysis (Quiroga et al., 2005)).”
- Could a Neuroscientist Understand a Microprocessor? (2016, background): “However, this may well be an illusion and many studies that have looked carefully at brain areas have revealed a daz-zling heterogeneity of responses [32, 33, 34].”
- Mechanisms of face perception. (2008, background): “Recordings from the medial temporal lobe of human epilepsy patients have revealed the existence of cells that respond to familiar individuals in a highly invariant manner (Quiroga et al 2005), as expected of a “grandmother cell”.”
- Coarse scales are sufficient for efficient categorization of emotional facial expressions: Evidence from neural computation (neutral): “This assumption is based in part on biological data reported in a single-cell recording study, which showed that neurons in the medial temporal lobe (MTL) respond independently to spatial location [30].”