The hippocampus, amygdala and entorhinal cortex receive convergent input from temporal neocortical regions specialized for processing complex visual stimuli and are important in the representation and recognition of visual images. Recording from 427 single neurons in the human hippocampus, entorhinal cortex and amygdala, we found a remarkable degree of category-specific firing of individual neurons on a trial-by-trial basis. Of the recorded neurons, 14% responded selectively to visual stimuli from different categories, including faces, natural scenes and houses, famous people and animals. Based on the firing rate of individual neurons, stimulus category could be predicted with a mean probability of error of 0.24. In the hippocampus, the proportion of neurons responding to spatial layouts was greater than to other categories. Our data provide direct support for the role of human medial temporal regions in the representation of different categories of visual stimuli.
Visual recognition of objects is a key function of the primate brain. There is a progression in the complexity of the representation of the visual scene by single neurons. Neurons in early visual areas in monkeys are tuned to simple features of the stimuli, such as the orientation of bars in area V1 or direction of motion in area V5. In the monkey inferotemporal cortex (IT), neurons respond to complex stimuli, including faces and hands, but also abstract patterns or common, everyday objects1,2,3. There are strong projections from IT to higher association areas in the temporal lobe, including the parahippocampal gyrus, perirhinal cortex, entorhinal cortex, hippocampus and amygdala4,5. Single neurons in these polymodal areas in monkeys also show visual selectivity for complex stimuli6,7,8.
Temporal lobe lesions lead to profound category-specific deficits in visual recognition in both macaques and humans9,10,11,12. There is evidence from electrical stimulation studies in epileptic patients that current injection in the temporal lobe can interfere with visual recognition13 and elicit visual memories and hallucinations13,14. Functional brain imaging and event-related potentials (ERP) also show a correlation between brain activity and visual recognition of specific categories of stimuli such as human faces and spatial layouts or places15,16,17,18,19,20.
We reported that neurons in the human medial temporal lobe discriminate objects from faces21. Here we further investigated visual response properties and showed that single neurons respond selectively to different stimulus categories.
We recorded the activity of 427 single neurons in 11 patients with pharmacologically resistant epilepsy who had intracranial depth electrodes implanted to determine the location of the seizure focus for possible surgical resection. Based on clinical criteria, electrode probes, containing several microwires each, were placed in medial temporal lobe targets bilaterally (Table 1). Based on MRI confirmation (Fig. 1), 149 of the neurons recorded were in the amygdala, 153 neurons in the entorhinal cortex and 125 neurons in the hippocampus. (Eighty-five percent of the sites were in the anterior segment of the hippocampus.) Most of the electrodes we recorded from were in the right temporal lobe (79%). During single-neuron recording, subjects were presented with visual stimuli (Fig. 2) and performed a simple discrimination task indicating whether the picture was a human face or not. Of the 427 neurons, 85 (20%) showed changes in firing rate during presentation of the visual stimuli, and 61 (14%) showed visually selective responses that were category specific.
Most neurons showed maintained firing rates below 10 spikes per second (Table 1). No significant response differences were observed between the right and left hemispheres, and therefore the data were pooled. The average overall firing rate was 3.6 ± 5.6 spikes per second, similar to observations in rats and monkeys6,7,8,22.
We studied the responses of each neuron to the 1000-ms presentation of the visual stimuli by averaging the activity for all images within each category. For each neuron and each category, a post-stimulus time histogram was computed, showing the neuronal response starting 1000 ms before stimulus onset and ending 1000 ms after the stimulus disappeared.
A neuron was considered visually selective for a specific category if the activity during stimulus presentation for that category was significantly different from the baseline activity and from the responses to other categories of stimuli (Methods). For example, a visually selective neuron in the entorhinal cortex had increased firing rate in response to pictures of animals (Fig. 3a). The activity of this neuron during the 100–1000 ms interval after stimulus onset was different from baseline for animal stimuli (p < 10−4, Wilcoxon rank-sum test), but not for the other stimulus categories ( p > 0.1). A one-way ANOVA comparing firing rates between categories yielded p < 0.001, and comparing the activity for animals to all other categories using a pair-wise non-parametric Wilcoxon test yielded statistically significant differences (p < 0.001). The latency of response for this neuron was 219 ms, and the duration of increased response over baseline was 752 ms.
How specific was the response of the neuron within the selective category? If the average increased response to animals were due to enhanced firing for only a few pictures of animals, one might expect to observe a bimodal (or multimodal) distribution of firing rates. However, the distribution of firing rates for this neuron during presentation of different pictures of animals did not show any clear signs of multimodality (Fig. 3b). Although there was variability in the response of the neuron to individual instances of animals, the neuron responded above baseline for all pictures of animals (Fig. 3c; p < 0.005 ). We also compared the variability for different presentations of the same animal to the variability across different pictures of animals using an analysis of variance (ANOVA). Both parametric and non-parametric tests failed to show differences among individual stimuli (p > 0.4).
One visually selective neuron in the anterior hippocampus (Fig. 4) showed an increased firing rate over baseline in response to drawings of famous people as well as, to a lesser degree, to photos of famous people (p < 0.001). A one-way ANOVA yielded p < 0.001, and subsequent across-categories, pair-wise comparisons also showed that the activity during stimulus presentation was significantly higher for these two categories. Although the peak response was larger for drawings than for photographs (13.9 spikes/s versus 9.6 spikes/s), the average activity was not significantly different (Wilcoxon rank-sum test, p > 0.05). This neuron did not respond to faces per se, as indicated by the lack of change in the activity for emotional faces of unknown actors. The distribution of firing rates in response to photos of famous people for this neuron did not show clear signs of displaying more than one mode (Fig. 4b). Variability in the responses to distinct individual photos (Fig. 4c) was not higher than variability across different presentations of the same photograph (ANOVA, p > 0.2). Similar results held for drawings of famous people. The response to all individual stimuli within the selective categories was significantly different from baseline (p < 0.01).
Although we used a significance criterion of 0.05 in the statistical analysis, most of the actual p values were below 0.01. For the visually selective neurons, 69% of the p values were less than 0.01 (average p value, 0.01 ± 0.02). A χ2 test rejected the hypothesis that these responses could be due to chance (p < 0.001).
Some of the neurons showed changes in firing rate in response to more than one of the categories (for example, Fig. 4). The proportion of neurons, relative to the number of selective neurons, that responded to more than one category was 21% in the amygdala, 10% in the entorhinal cortex and 25% in the hippocampus. Most of these neurons responded to two categories.
We also observed neurons that showed significant but non-selective changes in firing rate during stimulus presentation (Methods). The proportion of nonselective neurons, relative to the total number of responsive neurons, was 28% in the hippocampus, 29% in the entorhinal cortex and 17% in the amygdala.
Although most of the visually selective neurons showed increases in the firing rate on presentation of visual stimuli, some cells had reduced firing rate from baseline (three neurons in the amygdala, two in the entorhinal cortex and two in the hippocampus). Decreases were also observed in the visually responsive but nonselective neurons (one neuron in the amygdala, five in entorhinal cortex and three in hippocampus).
Most of the neurons responded during the stimulus-presentation period, but there were some that responded when the stimulus was removed. To address this, we computed for all neurons, within the 2000 ms after stimulus onset, the number of spikes in a 600-ms interval centered on the peak of the response and statistically analyzed the responses as described above. There were eight selective neurons (four in hippocampus, three in entorhinal cortex and one in the amygdala) that showed a statistically significant late response and were not detected with the previous analysis.
Under the assumption that there is no preference in the prevalence of selectivity for any of the nine different stimulus categories, the number of selective neurons within any one area should be uniformly distributed among these categories. We tested this hypothesis using a χ2 test23. We obtained χ2 values of 6.0, 9.6 and 29.5 for the amygdala, entorhinal cortex and hippocampus respectively (Fig. 5). The p values were over 0.25 for the amygdala and entorhinal cortex but less than 10−3 for the hippocampus. In the hippocampus, we observed a small relative proportion of responses to animals, food items and patterns and, interestingly, a relatively high number of neurons responding to spatial layouts (Fig. 5). No significant difference was observed in a direct comparison of the response of any of these neurons to house facades versus natural scenes (Wilcoxon, p > 0.05).
Among the selective neurons, there were 2 that yielded p < 0.05 (and 3 more with a p value between 0.05 and 0.1) in the ANOVA analysis of specificity to individual stimuli within the selective category. These neurons responded more strongly to one to three of the individual stimuli within the selective category. These were excluded from the number of selective neurons in Table 1.
We computed the latency and the duration of the evoked activities for all neurons showing a visual response (Methods). The latencies ranged from 52 to 695 ms and the durations from 53 to 1190 ms (Table 1). There was no significant difference among the amygdala, entorhinal cortex and hippocampus in either of these two variables (ANOVA, p > 0.1).
Classification by an ideal observer
We assessed how well an ideal observer could discriminate, based on the response of an individual neuron, whether a stimulus belonged to the category for which the neuron was selective or not. We computed pe, the probability of misclassifying a stimulus based on the firing rate, using a classical optimal decision procedure (ROC analysis; Methods and Fig. 6a–c). The value of pe can range from 0 to 0.5, with pe = 0.5 indicating chance performance and pe = 0 indicating perfect classification. Our pe values ranged from 0.13 to 0.32 (0.22 ± 0.06, mean ± s.d.) in the amygdala, 0.04 to 0.44 (0.23 ± 0.10) in the entorhinal cortex and 0.08 to 0.47 (0.23 ± 0.10) in the hippocampus (Fig. 6d–f).
Increasingly complex stimulus attributes are represented from the retina to the higher visual areas. Evidence from neurology11,12,24, functional brain imaging15,16,17,20 and evoked-potential studies in humans18, and from single-neuron electrophysiology1,2,3 and lesions in monkeys9,10,25 suggests a fundamental role for the medial temporal lobe in visual object recognition.
Category-specific knowledge deficits occur in which neurological patients show impairments in identifying living things, objects, food items or faces11,12,24. Functional imaging studies show activation of different areas in the temporal lobe that correlates with subjects' observing pictures belonging to different categories15,16,17. In particular, there are areas specialized for faces20,26, spatial layouts27,28, objects and animals17. Specific changes in activity on presentation of faces, objects and letter strings can also be observed by evoked potentials in humans29,30,31.
Single inferotemporal cortex (IT) neurons in monkeys respond to complex visual stimuli, including faces, objects and abstract patterns1,2,3. Neurons in human temporal neocortex respond to faces and words32,33. Information from these neocortical neurons is conveyed to polymodal association areas in the temporal lobe4,5.
We showed that in the relatively small matrix of hippocampus, entorhinal cortex and amygdala, there is a remarkable degree of segregation of categories at the level of single neurons. Neurons in these regions show visual object discrimination among at least nine stimulus categories. Based on the firing rate of individual neurons, it was possible to predict with a mean probability of error of 0.24 whether the preferred stimulus category was presented or not (Fig. 6). This, by itself, shows a striking degree of category-specific firing on a trial-by-trial basis. By combining the activity of multiple neurons, it is likely that an even higher level of accuracy can be achieved. Such category-specific processing may be important not only in object recognition, but also in the representation and retrieval processes that have been closely linked with the medial temporal lobe34,35,36.
Twenty percent of the neurons that we recorded showed a visual response and 14% a visually selective one. These percentages are comparable to those reported in monkeys. In the entorhinal cortex, 11% of the neurons show selective visual responses7, compared to 16% in our study. In the monkey amygdala, 12% of the neurons show visual responses and about 33% of those are selective for faces6, compared to 12% and 20%, respectively, in our data.
Most of the selective neurons responded to only a single stimulus category, rather than weakly responding to a large fraction of all stimuli. Our data thus support the existence of sparse coding in the medial temporal lobe. Sparsely coded representation has been suggested for information processing in the rodent and primate hippocampus37,38,39 and for processing of faces and objects in IT40,41.
Whereas a significant proportion of neurons are selective for faces in the superior temporal sulcus1,2,3, responses in the entorhinal cortex7, hippocampus8 and amygdala6 are much more varied. Responses were also diverse in our sample of entorhinal cortex and amygdala neurons (Fig. 5). However, in the hippocampus, we observed more responses to images showing spatial layouts, including houses, natural scenes and interiors. The rat hippocampus contains place cells that respond selectively to the position of a rat while it is navigating a maze22,42. Neurons in the monkey hippocampus respond selectively depending on the position of the stimulus in a conditioned spatial response task8. Functional MRI studies report parahippocampal43 and hippocampal44 activation associated with navigational tasks as well as while observing images similar to the ones shown in our study27,28. This area is posterior to the medial temporal recording sites in our study, and likely projects to the hippocampus.
It is possible that some neurons may change their activity more specifically than to the broad categories used in our study, responding, for instance, only to one specific example that we did not present45. Our experimental setting did not enable investigation of this issue for several reasons. First, because of clinical considerations, an electrode location remained fixed once placed, and we did not change electrode location in search for an optimal stimulus, as is commonly done in animal experiments. Also, because the data for all the channels were analyzed off-line, we could not determine, via immediate feedback, the ‘optimal’ stimulus for any of the cells. For most of the selective neurons (61 of 63), the analysis of variance showed that variability for different stimuli within the selective category was comparable to variability due to different presentations of the same stimulus.
Data from very different experiments and using distinct techniques are converging to show an important role for the human medial temporal lobe in visual object recognition. This study establishes that single neurons in humans explicitly respond to specific categories of stimuli, which may be relevant to the representation and retrieval of visual information.
Subjects were patients with pharmacologically resistant epilepsy. Extensive non-invasive evaluation did not yield concordant data corresponding to a single resectable epileptogenic focus, and therefore the patients were stereotactically implanted with up to 12 chronic intracranial depth electrodes for one to two weeks to determine the focus of their seizures for possible surgical resection21,46. Through the lumen of the electrodes, up to 8 microwires (40 μm diameter) were inserted. The surgeries were performed by I.F. All studies described here conformed with the guidelines of the Medical Institutional Review Board at UCLA. The present study describes data from 11 subjects (6 males and 5 females; 8 right-handed and 3 left-handed; 24 to 48 years old).
The sites of implantation of the electrodes were based exclusively on clinical criteria. The location of the electrodes was verified by structural MR images obtained before removing the electrodes (Fig. 1). Individual microwires extended approximately 4 mm from the tip, lying in a cone with an opening angle of less than 45 degrees.
We report here the activity of neurons for all the probes located in the medial temporal lobe. Neurons from anterior (85%), middle (10%) and posterior (5%) parts of the hippocampus were pooled together as hippocampus neurons. Most neurons in the amygdala were in the basolateral nuclear complex. Our MR resolution did not allow us to accurately determine in which CA fields the hippocampal probes were placed or what layer of the entorhinal cortex we recorded from.
The information recorded during seizures from the depth electrodes was used to localize the seizure focus46. Eighty-five percent of recorded neurons were outside the clinically determined zone of seizure onset (that is, either in the other hemisphere or in a different brain area on the same side). Ninety-four percent of responsive neurons were outside the seizure focus. Because we did not observe any differences in their waveforms, firing rates, interspike interval distributions or response properties, all neurons were included in Table 1. The percentage of responsive neurons would increase from 20% to 22% if we excluded neurons within the seizure focus.
A series of images was shown on a monitor at an approximate size of 5 degrees of visual angle (Fig. 2). Each picture was repeated 4–10 times (depending on time constraints), and there was a total of up to 600 presentations; the order of presentation of the stimuli was random. The number of different individual stimuli per category ranged from 3 to 25 (7.2 ± 4.6, mean ± s.d.). Each picture was presented for 1000 ms. In the first two patients, stimuli from only three different categories were presented (emotional faces, objects and spatial layouts). Immediately after the picture disappeared, there was a tone that indicated that the subject had to respond whether the picture was a human face or not by pressing a button. This was done to engage the subject's attention and to verify that he was seeing the pictures. Trials in which subjects made an error (mean percentage correct 97 ± 1%) or in which the subjects moved were discarded from subsequent analysis. The behavioral response occurred on average 472 ± 201 ms after stimulus offset.
Data from each of the recorded microwires were amplified and high-pass filtered (with a corner frequency of 300 Hz), A/D converted and stored for off-line spike sorting using Experiment Workbench data acquisition software (Datawave, Denver, Colorado). Note that because the microelectrodes were chronically implanted, we could not further select neurons by moving the electrodes, as is common in monkey experiments. We analyzed all spikes from all neurons we detected.
Spikes from single neurons were discriminated from the extracellular recordings based on the height, width, peak voltage and other parameters of the waveforms using a manual cluster-cutting method implemented in Datawave. For each isolated neuron, we determined the fraction of all spikes that were within 2 ms of each other. If this fraction exceeded 2%, the data were discarded because of possible contamination by firing from more than one neuron.
In some cases, we recorded from the same microwires on separate days (presenting different sets of pictures from the same categories). Because of various constraints, we often could not be sure that a neuron recorded on one day was identical to a neuron recorded on another day. Of our 427 neurons, 128 (30%) were recorded on consecutive days. If we assume that all 128 neurons are identical across days, the total number of neurons reduces to 299, and the percentage of responsive neurons increases from 20% (85 of 427) to 23% (70 of 299).
For each neuron, we computed histograms locked to the stimulus presentation times (post-stimulus time histograms, PSTHs) by averaging the neuronal responses for all stimuli within each category. We also compared all color versus non-color pictures; we found only two neurons with a statistically significant enhanced response to color stimuli. To assess the significance of the response to different categories of visual stimuli, we ran a series of statistical comparisons. For a neuron to be considered visually selective for a specific category, it had to meet three requirements. First, the neuronal response pooled over all stimuli belonging to the same category had to show a different firing rate, that is, lower or higher, during the time of presentation of the image than during the preceding baseline. The firing rate during image presentation was computed in the interval 100 ms ≤ t < 1000 ms. (The lower boundary of 100 ms was chosen because most neurons showed latencies above 100 ms.) The baseline before stimulus presentation was computed in the interval −1000 ms ≤ t < 0 ms. Significance was assessed both by a two-tailed t-test and a non-parametric Wilcoxon rank-sum test23. Because the results from both tests were very similar, the p values reported throughout the text for pairwise comparisons correspond to those from the non-parametric test. Second, a one-way analysis of variance (ANOVA) during the 100–1000 ms interval after stimulus onset, assessing whether there were differences in firing rate among the different categories, had to show a significant p-value (< 0.05). In this analysis, stimuli were pooled according to the category they belonged to. Third, subsequent pairwise comparisons between the activity during this interval for the putative selective category and the rest of the categories had to show a statistically significant change (Wilcoxon rank sum test, < 0.05).
Although we used a significance criterion of 0.05, most responsive neurons showed p values that were less than 0.01 (Results ). Our analysis encompassed both increases and decreases in firing rate.
If the ANOVA test failed to indicate a significant difference between categories, but the activity for 75% of the stimulus categories was significantly different from the baseline activity (that is, the first criterion was met but not the second), we labeled the neuron as visually responsive but nonselective.
For the neurons that showed visual selectivity, we further studied the degree of specificity of the responses. We computed the distribution of firing rates for the selective category to assess whether it was a polymodal distribution. We also computed an analysis of variance (both a parametric analysis23 as well as a non-parametric analysis using a bootstrap procedure47) to compare the variance to different individual stimuli within the same category to the variance to repeated presentations of the same stimulus.
It is reported that neurons can respond to visual stimuli with a long delay even after the stimulus disappears21,48,49. To study these cases, we analyzed the responses in an interval of 600 ms centered on the response peak. To estimate the time of occurrence of the peak, we computed an estimation of the spike density function (sdf) for each category of stimuli by convolving the spike trains with a Gaussian of fixed width of 100 ms and then averaging over repetitions.
We computed the latency and duration of the responses from the sdf for each category. The latency was defined as the first time value on which five consecutive bins of the sdf yielded a value that was beyond two s.d. of the mean response before stimulus presentation. In analogous fashion, the end of the response was defined as the first time point at which five consecutive bins of the sdf yielded a value not different from baseline after the latency value.
To discriminate whether the neurons showed significant activity related to the behavioral response, we did two additional analyses. First, we computed histograms of all the neuronal responses locked to the time at which the subjects pressed the button by averaging over all stimuli. We then computed whether there was a significant difference in the response in the interval 300 ms before and 300 ms after the button was pressed. This interval was chosen so that it would not overlap on average with the periods of visual presentation. We found four neurons (three in the amygdala, one in the hippocampus) for which the response before the button press was significantly different from that after the button press and also from that during the baseline −1000 to 0 ms interval (1-way ANOVA, p < 0.05; pairwise comparisons, p < 0.05). None of these neurons was visually responsive. We also found 2 neurons that showed an increased activity in a 600-ms window around the response compared to the baseline (Wilcoxon test, p < 0.05). Neither of these neurons was visually responsive.
When computing the average neuron activity in response to all human faces (emotional faces, drawings and photos of famous faces), we observed 10 neurons (5 in the hippocampus, 3 in the entorhinal cortex, 2 in the amygdala) for which the activity in the 100–1000 ms interval for all face stimuli was significantly larger than the baseline activity and than the response for non-face stimuli (1-way ANOVA, p < 0.05; pairwise comparisons, p < 0.05). However, for some of these neurons (for example, Fig. 4a), the response was selective for some faces and not others.
For those neurons that showed visual selectivity for stimuli within specific categories, we also addressed the question of how well an ideal observer could estimate which category the stimulus belonged to by observing the firing rate. To quantify the classification performance of the neurons, we used an ROC analysis as used in signal-detection theory and psychophysics experiments50. For each visually selective neuron, we computed the distribution of firing rates for the preferred stimuli and the non-preferred stimuli (the remaining categories; Fig. 6a). From the distribution of firing rates we evaluated, by sliding a threshold T over the whole range of firing rates, the probability of correct detection (PCD) and the probability of false alarm (PFA). Assuming chance performance, PCD = PFA (Fig. 6b, dashed line). The departure from the diagonal shows the possibility of discriminating between the preferred and non-preferred categories. The probability of misclassification plotted against the probability of false alarm can be obtained as
The overall probability of error, pe, is then defined as the minimum value of this function (arrow in Fig. 6c). A value of pe = 1/2 corresponds to chance performance, whereas a value of pe = 0 indicates that it is possible to predict with perfect accuracy based on the number of spikes whether the specified category was presented or not.
Throughout the manuscript, values are given as mean ± standard deviation (s.d.).
Logothetis, N. K. & Sheinberg, D. L. Visual object recognition. Annu. Rev. Neurosci. 19, 577 –621 (1996).
Tanaka, K. Inferotemporal cortex and object vision. Annu. Rev. Neurosci. 19, 109–139 (1996).
Gross, C. G. How inferior temporal cortex became a visual area. Cereb. Cortex 5, 455–469 (1994).
Suzuki, W. A. Neuroanatomy of the monkey entorhinal, perirhinal and parahippocampal cortices: Organization of cortical inputs and interconnections with amygdala and striatum . Sem. Neurosci. 8, 3–12 (1996).
Saleem, K. S. & Tanaka, K. Divergent projections from the anterior inferotemporal area TE to the perirhinal and entorhinal cortices in the macaque monkey. J. Neurosci. 16, 4757– 4775 (1996).
Leonard, C. M., Rolls, E. T., Wilson, F. A. W. & Baylis, G. C. Neurons in the amygdala of the monkey with responses selective for faces. Behav. Brain. Res 15, 159–176 (1985).
Suzuki, W., Miller, E. & Desimone, R. Object and place memory in the macaque entorhinal cortex . J. Neurophysiol. 78, 1062– 1081 (1997).
Miyashita, Y., Rolls, E. T., Cahusac, P. M. B., Niki, H. & Feigenbaum, J. D. Activity of hippocampal formation neurons in the monkey related to a conditional spatial response task. J. Neurophysiol. 61, 669–678 (1989).
Mishkin, M. Memory in monkeys severely impaired by combined but not separate removal of amygdala and hippocampus. Nature 273, 297 –298 (1978).
Meunier, M., Bachevaoer, J., Mishkin, M. & Murray, E. A. Effects on visual recognition of combined and separate ablations of the entorhinal and perirhinal cortex in rhesus monkeys. J. Neurosci. 13, 5418–5432 (1993).
Tranel, D., Damasio, H. & Damasio, A. Face agnosia and the neural substrates of memory. Annu. Rev. Neurosci. 13, 89–109 (1990).
Warrington, E. & McCarthy, R. Categories of knowledge — Further fractionations and an attempted integration. Brain 110, 1273–1296 (1987).
Fried, I., Mateer, C., Ojemann, G., Wohns, R. & Fedio, P. Organization of visuospatial functions in human cortex . Brain 105, 349–371 (1982).
Penfield, W. & Jasper, H. Epilepsy and the Functional Anatomy of the Human Brain (Little, Brown, Boston, 1954).
Treisman, A. M. & Kanwisher, N. G. Perceiving visually presented objects: recognition, awareness, and modularity. Curr. Opin. Neurobiol. 8, 218–226 (1998).
Ishai, A., Ungerleider, L. G., Martin, A., Schouten, J. L. & Haxby, J. V. Distributed representation of objects in the human ventral visual pathway. Proc. Natl. Acad. Sci. USA 96, 9379–9384 (1999).
Martin, A., Wiggs, C. L., Ungerleider, L. G. & Haxby, J. V. Neural correlates of category-specific knowledge. Nature 379, 649–652 (1996).
Allison, T. et al. Face recognition in human extrastriate cortex. J. Neurophysiol. 71, 821–825 (1994).
Nobre, A., Allison, T. & McCarthy, G. Word recognition in the human inferior temporal lobe . Nature 372, 260–263 (1994).
Tong, F., Nakayama, K., Moscovitch, M., Weinrib, O. & Kanwisher, N. Response properties of the human fusiform face area. Cognit. Neuropsychol. 17, 257–279 (2000).
Fried, I., MacDonald, K. A. & Wilson, C. Single neuron activity in human hippocampus and amygdala during recognition of faces and objects. Neuron 18, 753–765 (1997).
Wilson, M. A. & McNaughton, B. L. Dynamics of the hippocampal ensemble code for space. Science 261, 1055 –1058 (1993).
Keeping, E. S. Introduction to Statistical Inference (Dover, New York, 1995).
Adolphs, R., Tranel, D., Damasio, H. & Damasio, A. Impaired recognition of emotion in facial expressions following bilateral damage to the amygdala . Nature 372, 669–672 (1994).
Buckley, M. J. & Gaffan, D. Perirhinal cortex ablation impairs object identification. J. Neurosci. 18, 2268–2275 (1998).
Kanwisher, N., McDermott, J. & Chun, M. M. The fusiform face area: a module in human extrastriate cortex specialized for face perception. J. Neurosci. 17, 4302–4311 (1997).
Epstein, R. & Kanwisher, N. A cortical representation of the local visual environment. Nature 392, 598 –601 (1998).
Epstein, R., Harris, A., Stanley, D. & Kanwisher, N. The parahippocampal place area: recognition, navigation, or encoding? Neuron 23, 115–125 (1999).
Allison, T., Puce, A., Spencer, D. & McCarthy, G. Electrophysiological studies of human face perception. I: Potentials generated in occipitotemporal cortex by face and non-face stimuli. Cereb. Cortex 9, 415–430 (1999).
Klopp, J., Halgren, E., Marinkovic, K. & Nenov, V. Face-selective spectral changes in the human fusiform gyrus. Clin. Neurophysiol. 110, 676–682 (1999).
Halgren, E., Raij, T., Marinkovic, K., Jousmaki, V. & Hari, R. Cognitive response profile of the human fusiform face area as determined by MEG. Cereb. Cortex 10, 69–81 (2000).
Ojemann, J. G., Ojemann, G. A. & Lettich, E. Neuronal activity related to faces and matching in human right nondominant temporal cortex. Brain 115, 1–13 (1992).
Ojemann, G. A., Ojemann, S. G. & Fried, I. Lessons from the human brain: neuronal activity related to cognition. The Neuroscientist 4, 285– 300 (1998).
Scoville, W. B. & Milner, B. Loss of recent memory after bilateral hippocampal lesions. J. Neurol. Neurosurg. Psychiatry. 20, 11–21 (1957).
Penfield, W. & Milner, B. Memory deficit produced bilateral lesions in the hippocampal zone. Arch. Neurol. Psychol. 79, 475–479 (1958).
Zola-Morgan, S. & Squire, L. R. Neuroanatomy of memory. Annu. Rev. Neurosci. 16, 547– 563 (1993).
McNaughton, B. L. & Nadel, L. in Neuroscience and Connectionist Theory (ed. Seifert, W.) 1– 64 (Academic, New York, 1990).
Alvarez, P. & Squire, L. R. Memory consolidation and the medial temporal lobe: a simple network model. Proc. Natl. Acad. Sci. USA 91, 7041–7045 (1994).
Treves, A. & Rolls, E. T. Computational analysis of the role of the hippocampus in memory. Hippocampus 4, 374–391 (1994).
Young, M. P. & Yamane, S. Sparse population coding of faces in the inferior temporal cortex. Science 256, 1327–1331 (1992).
Logothetis, N. K. & Pauls, J. Psychophysical and physiological evidence for viewer-centered object representations in the primate . Cereb. Cortex 3, 270– 288 (1995).
Muller, R. A quarter of a century of place cells. Cell 17, 813–822 (1996).
Aguirre, G. K., Detre, J. A., Alsop, D. C. & D'Esposito, M. The parahippocampus subserves topographical learning in man. Cereb. Cortex 6, 823–829 (1996).
Maguire, E. A. et al. Knowing where and getting there: a human navigation network . Science 280, 921–924 (1998).
Heit, G., Smith, M. E. & Halgren, E. Neural encoding of individual words and faces by the human hippocampus and amygdala. Nature 333, 773–775 (1988).
Fried, I. et al. Cerebral microdialysis combined with single-neuron and electroencephalographic recording in neurosurgical patients. J. Neurosurg. 91, 697–705 (1999).
Efron, B. & Tibshirani, R. J. An Introduction to the Bootstrap (Chapman & Hall/CRC, London, 1993).
Yakovlev, V., Fusi, S., Berman, E. & Zohary, E. Inter-trial neuronal activity in inferior temporal cortex: a putative vehicle to generate long-term visual associations. Nat. Neurosci. 1, 310 –317 (1998).
Miyashita, Y. Neuronal correlate of visual associative long-term memory in the primate temporal cortex. Nature 335, 817– 820 (1988).
Green, D. & Swets, J. Signal Detection Theory and Psychophysics (Wiley, New York, 1966).
This work was supported by grants from NIH (to I.F.), the Keck Foundation (to C.K.) and the Center for Consciousness Studies, University of Arizona (to I.F. and C.K.). We would like to thank D. Rozenfarb and M. Zirlinger for suggestions and reading the manuscript and Peter Steinmetz for general discussion. We wish to acknowledge Tony Fields, Jack Morrow, Eve Isham, Charles Wilson, Rick Staba, Eric Behnke and Anatol Bragin for help with the recordings.
About this article
Cite this article
Kreiman, G., Koch, C. & Fried, I. Category-specific visual responses of single neurons in the human medial temporal lobe. Nat Neurosci 3, 946–953 (2000). https://doi.org/10.1038/78868
Translational Psychiatry (2021)
Frontiers in Human Neuroscience (2020)
Scientific Reports (2020)