Epileptic Disorders


Investigating temporal pole function by functional imaging Volume 4, supplément 1, Supplement 1, September 2002

In 1930, Klüver and Bucy [1] reported that bilateral ablation of the temporal lobe in monkeys induced abnormal social and sexual behavior as well as a deficit in visual cognitive function. In the early 1950's, Chow [2] performed more selective temporal lobectomy in monkeys involving ablation of just the anterior part of the lobe including the temporal pole. This resulted in a deficit in food recognition and compromised the animals' performance in visual pattern discrimination tests. More recently, Mishkin [3] reported that ablation of just the temporal pole and the amygdala resulted in changes in oral and visual object recognition in monkeys. However, none of these studies were confined to the temporal pole and the possibility that other, adjacent structures (i.e. the amygdala, the entorhinal cortex and the temporal convolutions) had been damaged meant that it was impossible to ascribe any particular function to the temporal pole region. In the second half of the twentieth century, the specific role of the temporal pole was investigated by comparing anatomical and clinical data in humans, and by means of experiments conducted in animals, mainly primates. The techniques of functional imaging ­ notably positron emission tomography (PET) and magnetic resonance imaging (MRI) ­ have significantly advanced our understanding of temporal pole function. Initially, this type of approach was simply used to confirm prior observations made in humans and animals, but more recently other functional roles have been brought to light in both healthy subjects and patients with epilepsy, dementia and psychosis.

The role of the temporal pole The traditional approach: data from human and animal studies

Anatomo-clinical correlations in humans

In the literature, there are very few "pure" anatomo-clinical correlations which cast light on temporal pole function. Most reports concern the involvement of this region in autobiographical memory. For example, Kapur et al. [4] present the case history of a young patient who, following a riding accident, presented a deficit in recognition of familiar scenes, faces and names as well as amnesia for the period preceding the cranial injury. The patient's MRI scan showed bilateral damage to area 38 and the anterior parts of areas 21 and 28, i.e. more or less exclusively involving the temporal pole. Another interesting case, reported by Papagno and Capitani [5], concerned a patient who had progressive cerebral atrophy mainly involving the left temporal pole. Initially, this patient had presented with severe amnestic aphasia for the names of her family members and friends without any concomitant impairment of semantic knowledge or of her understanding of these names. Later, the anomia deepened to include common nouns.

Animal experiments

Most of the relevant experimental data derived from animals were obtained in lesional studies in primates. In 1957, Pinto-Hamuy et al. [6] reported that bilateral temporal pole ablation in monkeys induced food recognition deficits ­ visual but also olfactory and gustatory. Horel et al. [7] later demonstrated that ablation of the cortex in the ventral and anterior temporal pole resulted in an inability to recognize faces. Gaffan [8] revealed the role played by the temporal pole in the visual discrimination of two-dimensional patterns: in a test based on discrimination with respect to complex visual scenes, two images (one positive and the other negative) were presented to monkeys on a screen. The animals had been conditioned to select the positive image but after temporal pole ablation were incapable of making the correct choice. This confirms the importance of the role of the temporal polar cortex ­ and more specifically of the anterior ventromedial area ­ in the visual discrimination of two-dimensional patterns and face recognition. Later, various authors [8-10] further elucidated the role played by the temporal pole in matching and learning tasks.

The role of the temporal pole The functional imaging approach

Confirmation of anatomo-clinical correlations in humans

In 1999, Maguire and Mummery [11] confirmed the importance of the temporal pole in autobiographical memory function in a study in which eight volunteer subjects were examined by positron emission tomography. Various memory-related parameters (including both time-dependent and time-independent parameters) were assessed: memory of autobiographical events, of public events, of autobiographical facts and general knowledge. They demonstrated that a specific cerebral network became activated during performance of recognition tasks, whatever the type of memory being investigated. This network involved: the medial prefrontal cortex (Brodmann's area 10), the left anterolateral middle temporal gyrus (Brodmann's area 21), the left temporal pole (Brodmann's area 38), the hippocampus, the left hippocampal gyrus, the posterior cingulate cortex (Brodmann's area 31), and both the left and right temporoparietal junctions (Brodmann's area 39). Different parts of this network were differentially activated in different memory tasks, e.g. remembering of time-dependent autobiographical events specifically activated the left temporal pole region, the left hippocampus and the medial prefrontal region thereby confirming the specific role played by the temporal pole (in this case the left temporal pole) in autobiographical memory.

Confirmation of data obtained in animal experiments

Sergent et al. [12] used PET in healthy volunteers to confirm that the temporal pole plays a role in face recognition (Figure 1). Of the six discrimination and recognition tasks performed (including discrimination between types, semantic categories and faces), the left temporal pole region was selectively activated only during the face recognition task. Similarly, Vandenberghe et al. [13] using the same methodology observed significant reduction in cerebral blood flow in the medial region of the left temporal pole during a task involving visual discrimination between different two-dimensional geometric patterns. The reduction in polar cortex activity was proportional to the familiarity with the geometric shapes presented, as had been previously observed in monkeys (Figure 2). Finally, another PET study of cerebral blood flow by Roland et al. [14] confirmed that the temporal pole is active in mnemonic matching and learning tasks.

New roles of the temporal pole identified by experiments in healthy subjects

One of the contributions of functional imaging is the discovery of the role played by the temporal cortex in language processing [15]. Activation of this region ­ both the right and the left ­ during linguistic tasks has been observed in a number of studies. Maguire et al. [16] studied cerebral activation patterns associated with reading stories of varying degrees of comprehensibility and consistency (although all were written in a grammatically correct style). The left temporal pole was shown to be preferentially activated by overall understanding, more specifically in establishment of the key narrative links in the story. The authors advanced the hypothesis that the left temporal pole is involved in linguistic integration processes which make it possible to create semantic and lexical links between words to get an overall understanding of a story without any reference to preexisting knowledge. A number of other articles [17, 18] have similarly described preferential activation of the right temporal pole when handling stories with emotional or affective content. No such activation was observed if the story was presented in other than the subject's maternal language. This led to the concept that the right temporal pole is involved in integrating the emotional content of words and stories.

New functions of the temporal pole identified by experiments in patients

The temporal lobe epilepsy model

In a PET study of cerebral metabolism by Semah et al., correlation emerged between the level of metabolic activity in the hippocampal-polar region and mnemonic scores for immediate and deferred memory tasks. Ten patients suffering from left mesial temporal lobe epilepsy with ipsilateral hippocampal sclerosis were included in this study. All were asked to perform mnemonic tests which involved learning a list of thirteen abstract words [19] followed by immediate recall after 30 min and deferred recall after twenty-four hours. The mnemonic score was expressed in the number of the words recalled correctly. In a second stage of the study, the same patients were examined using FDG-PET to assess cerebral metabolic activity. Results showed a strong correlation between left hippocampal-polar metabolism and mnemonic scores for deferred memory (Figure 3). There was no correlation between any mnemonic performance and metabolic activity in either the polar or hippocampal regions alone, thus suggesting that it may be the temporal hippocampal-polar complex which is involved in episodic verbal memory recall.

The Alzheimer's disease model

Pathological examinations have revealed [20] that Alzheimer's disease-specific lesions (senile plaques, neurofibrillary tangles and neuronal loss) can be found in molecular layers III and V of the cortex of the temporal pole. Since the temporal pole cortex is intimately interconnected with the sensory association areas and limbic areas responsible for memory, the hypothesis was proposed that polar damage might contribute to the cognitive deficit and behavioral problems seen in patients with Alzheimer's disease. In order to investigate this hypothesis, Ouchi et al. [21] performed FDG-PET to compare cerebral metabolism in three different groups; healthy volunteers, amnesic patients with intact higher functions, and Alzheimer's patients. In the Alzheimer's group, reduced metabolic activity was observed in the head of the hippocampus and the amygdala, to a lesser extent in the lateral temporal cortex and, more specifically, in the temporal pole. Such temporal pole hypometabolism was not observed in either healthy controls or purely amnesic patients. This provided evidence that damage to the amygdalo-hippocampal complex and to the adjacent connecting cortex is a key factor in the etiology of the cognitive deficit and behavioral problems seen in patients with Alzheimer's disease.

The schizophrenia model

Morphological imaging techniques have been applied comparing the amounts of white and gray matter in schizophrenics [22]. It has been shown that such patients have significantly less gray matter in both right and left temporal poles, although no correlation could be established between the precise localization of this abnormality and specific symptoms.

Here again, it seems that the temporal pole is specifically and intimately involved in a pathological process.


Temporal pole function has been a neglected topic but it is now becoming apparent that this region plays a key role in various important functions, including autobiographical memory, face and visual pattern recognition, mnemonic matching and learning tasks, linguistic integration and the processing of emotional language; possibly also in episodic verbal memory recall. Future work will define if these functions are handled by the temporal pole itself or are managed in a more extensive, and complex neuronal network, as seems likely given the numerous anatomical connections to and from the temporal pole which are known to exist.