Face Recognition Units Research Articles

Self- or familiar-face recognition advantage? New insight using ambient images.

Self-face recognition has been widely explored in the past few years. Nevertheless, the current literature relies on the use of standardized photographs which do not represent daily-life face recognition. Therefore, we aim for the first time to evaluate self-face processing in healthy individuals using natural/ambient images which contain variations in the environment and in the face itself. In total, 40 undergraduate and graduate students performed a forced delayed-matching task, including images of one's own face, friend, famous and unknown individuals. For both reaction time and accuracy, results showed that participants were faster and more accurate when matching different images of their own face compared to both famous and unfamiliar faces. Nevertheless, no significant differences were found between self-face and friend-face and between friend-face and famous-face. They were also faster and more accurate when matching friend and famous faces compared to unfamiliar faces. Our results suggest that faster and more accurate responses to self-face might be better explained by a familiarity effect - that is, (1) the result of frequent exposition to one's own image through mirror and photos, (2) a more robust mental representation of one's own face and (3) strong face recognition units as for other familiar faces.

Evidence for view-invariant face recognition units in unfamiliar face learning.

Many models of face recognition incorporate the idea of a face recognition unit (FRU), an abstracted representation formed from each experience of a face which aids recognition under novel viewing conditions. Some previous studies have failed to find evidence of this FRU representation. Here, we report three experiments which investigated this theoretical construct by modifying the face learning procedure from that in previous work. During learning, one or two views of previously unfamiliar faces were shown to participants in a serial matching task. Later, participants attempted to recognize both seen and novel views of the learned faces (recognition phase). Experiment 1 tested participants' recognition of a novel view, a day after learning. Experiment 2 was identical, but tested participants on the same day as learning. Experiment 3 repeated Experiment 1, but tested participants on a novel view that was outside the rotation of those views learned. Results revealed a significant advantage, across all experiments, for recognizing a novel view when two views had been learned compared to single view learning. The observed view invariance supports the notion that an FRU representation is established during multi-view face learning under particular learning conditions.

A High-Efficiency and High-Accuracy Fully Automatic Collaborative Face Annotation System for Distributed Online Social Networks

The development of fully automatic face annotation techniques in online social networks is currently very important for effective management and organization of a large number of personal photos shared on social network platforms. In this paper, we first propose the personalized hierarchical database access architecture for each member by taking advantage of various social network context types to substantially reduce time consumption. Next, we construct the personalized and adaptive fused face recognition (FR) unit for each member, which uses the AdaBoost algorithm to fuse several different types of base classifiers to produce highly reliable face annotation results. Additionally, to efficiently select suitable personalized face recognizers and then effectively merge multiple personalized face recognizer results, we propose two collaborative FR strategies: the owner with a highest priority rule and using a weighted majority rule for query photos within our collaborative FR framework. The experiment results demonstrate that the evaluation methodologies produced F -measure and Similarity accuracy rates that were, respectively, 64.03% and 63.05% higher for the proposed method in comparison to other state-of-the-art face annotation methods, as well as demonstrating that our method can result in a reduction in overall processing time of 78.06%.

The Neuropsychology of Familiar Person Recognition from Face and Voice

Prosopagnosia has been considered for a long period of time as the most important and almost exclusive disorder in the recognition of familiar people. In recent years, however, this conviction has been undermined by the description of patients showing a concomitant defect in the recognition of familiar faces and voices as a consequence of lesions encroaching upon the right anterior temporal lobe (ATL). These new data have obliged researchers to reconsider on one hand the construct of ‘associative prosopagnosia’ and on the other hand current models of people recognition. A systematic review of the patterns of familiar people recognition disorders observed in patients with right and left ATL lesions has shown that in patients with right ATL lesions face familiarity feelings and the retrieval of person-specific semantic information from faces are selectively affected, whereas in patients with left ATL lesions the defect selectively concerns famous people naming. Furthermore, some patients with right ATL lesions and intact face familiarity feelings show a defect in the retrieval of person-specific semantic knowledge greater from face than from name. These data are at variance with current models assuming: (a) that familiarity feelings are generated at the level of person identity nodes (PINs) where information processed by various sensory modalities converge, and (b) that PINs provide a modality-free gateway to a single semantic system, where information about people is stored in an amodal format. They suggest, on the contrary: (a) that familiarity feelings are generated at the level of modality-specific recognition units; (b) that face and voice recognition units are represented more in the right than in the left ATLs; (c) that in the right ATL are mainly stored person-specific information based on a convergence of perceptual information, whereas in the left ATLs are represented verbally-mediated person-specific information.

Mental representations of familiar faces

The Bruce and Young (1986) framework makes a number of important distinctions between the types of representation needed to recognize a familiar face. Here, we return to these, focussing particularly on face recognition units. We argue that such representations need to incorporate idiosyncratic within-person variability, asking questions such as 'What counts as a picture of Harrison Ford?'. We describe a mechanism for achieving this, and discuss the relation between image variability and episodic face memories, in the context of behavioural and neurophysiological data.

Multiple contributions to priming effects for familiar faces: Analyses with backward masking and event‐related potentials

The model of face recognition by Bruce and Young postulates a pool of structural representations for familiar faces in long-term memory, so-called face recognition units (FRUs). Event-related brain potentials show early repetition priming effects for familiar faces around 250-300 ms [N250r or early repetition effect (ERE)], which are thought to reflect the activation of these FRUs. However, small N250r effects are also seen for unfamiliar faces suggesting that priming of perceptual codes (i.e., pictorial and structural codes) also contributes to early repetition effects. Using a face-familiarity task in Experiment 1, we aimed to eliminate these perceptual contributions to face priming by backward masking the prime face with a different, unfamiliar face. As expected, a repetition priming effect appeared only for familiar faces. Experiment 2 used a semantic-decision task and compared the effects of different kinds of masks that interfered with either pictorial codes or with pictorial and structural codes. Our findings indicate that both structural codes and memory representations contribute to the N250r and that unfamiliar-face masks interfere only with structural codes. Face-masks may therefore provide a useful tool to extract the pure contributions of memory representations (i.e., FRUs) to repetition priming.

Identity-specific face adaptation effects: Evidence for abstractive face representations

The effects of selective adaptation on familiar face perception were examined. After prolonged exposure to photographs of a celebrity, participants saw a series of ambiguous morphs that were varying mixtures between the face of that person and a different celebrity. Participants judged fewer of the morphs to resemble the celebrity to which they had been adapted, implying that they were now less sensitive to that particular face. Similar results were obtained when the adapting faces were highly dissimilar in viewpoint to the test morphs; when they were presented upside-down; or when they were vertically stretched to three times their normal height. These effects rule out explanations of adaptation effects solely in terms of low-level image-based adaptation. Instead they are consistent with the idea that relatively viewpoint-independent, person-specific adaptation occurred, at the level of either the “Face Recognition Units” or “Person Identity Nodes” in Burton, Bruce and Johnston’s (1990) model of face recognition.

Associated impairment of the categories of conspecifics and biological entities: Cognitive and neuroanatomical aspects of a new case

Case A.C.A. presented an associated impairment of visual recognition and semantic knowledge for celebrities and biological objects. This case was relevant for (a) the neuroanatomical correlations, and (b) the relationship between visual recognition and semantics within the biological domain and the conspecifics domain. A.C.A. was not affected by anterior temporal damage. Her bilateral vascular lesions were localized on the medial and inferior temporal gyrus on the right and on the intermediate fusiform gyrus on the left, without concomitant lesions of the parahippocampal gyrus or posterior fusiform. Data analysis was based on a novel methodology developed to estimate the rate of stored items in the visual structural description system (SDS) or in the face recognition unit. For each biological object, no particular correlation was found between the visual information accessed through the semantic system and that tapped by the picture reality judgement. Findings are discussed with reference to whether a putative resource commonality is likely between biological objects and conspecifics, and whether or not either category may depend on an exclusive neural substrate.

Do I Know You? Insights into Memory for Faces from Brain Potentials

The recognition of faces is central to human social interaction. Recordings of event-related potentials (ERPs) from the brain can shed light on the various processes that occur when a face is recognized and when knowledge related to a specific person is retrieved. ERP contrasts between processing familiar and processing novel faces offer a gateway into investigations of semantic memory for familiar persons. In particular, activity of face recognition units and semantic information units--memory representations of faces and person-related knowledge, respectively--can be indexed by specific ERPs. These potentials thus provide valuable tools for studying the cognitive and neurobiological architecture of person recognition. ERPs have also been found useful for investigating other types of memory for faces. Specifically, important insights have been derived from the study of a category of memory phenomena known as priming. Priming can be revealed in special tests when face recognition is facilitated based on prior experience. Describing the neural processes associated with memory for faces is an exciting focus of research, and future results from this line of inquiry promise to provide further knowledge about face recognition and the various types of memory that can be provoked by a human face.

Hyperfamiliarité stéréotypée pour les visages au cours d’une démence fronto-temporale avec prosopagnosie

The association of prosopagnosia and false recognition of faces is unusual and contributes to our understanding of the generation of facial familiarity. A 67-year-old man with a left prefrontal traumatic lesion, developed a temporal variety of fronto-temporal dementia (semantic dementia) with amyotrophic lateral sclerosis. Cerebral imagery demonstrated a bilateral, temporal anterior atrophy predominating in the right hemisphere. The main cognitive signs consisted in severe difficulties to recognize faces of familiar people (prosopagnosia), associated with systematic false recognition of unfamiliar people. Neuropsychological testing indicated that the prosopagnosia probably resulted from the association of an associative/mnemonic mechanism (inability to activate the Face Recognition Units (FRU) from the visual input) and a semantic mechanism (degradation of semantic/biographical information or deconnexion between FRU and this information). At the early stage of the disease, the patient could activate residual semantic information about individuals from their names, but after a 4-year course, he failed to do so. This worsening could be attributed to the extension of the degenerative lesions to the left temporal lobe. Familiar and unfamiliar faces triggered a marked feeling of knowing. False recognition concerned all the unfamiliar faces, and the patient claimed spontaneously that they corresponded to actors, but he could not provide any additional information about their specific identities. The coexistence of prosopagnosia and false recognition suggests the existence of different interconnected systems processing face recognition, one intended to identification of individuals, and the other producing the sense of familiarity. Dysfunctions at different stages of one or the other of these two processes could result in distortions in the feeling of knowing. From this case and others reported in literature, we propose to complete the classical model of face processing by adding a pathway linked to limbic system and frontal structures. This later pathway could normally emit signals for familiarity, essentially autonomic, in response to the familiar faces. These signals, primitively unconscious, secondly reach consciousness and are then integrated by a central supervisor system which evaluates and verifies identity-specific biographical information in order to make a decision about the sense of familiarity.

Perceptual fluency, semantic familiarity and recognition-related familiarity: an electrophysiological exploration

Scalp recorded event-related potentials (ERPs) were used to examine the neuronal activity associated with perceptual fluency, semantic familiarity and recognition-related familiarity. We assume that ERP differences between first and second presentations of non-famous faces in an implicit memory condition reflect perceptual fluency, ERP differences between first presentations of famous and non-famous faces reflect semantic familiarity (i.e., familiarity arising from semantic memory retrieval), and early ERP differences between first and second presentations of non-famous and famous faces in an explicit recognition memory task reflect recognition-related familiarity. Semantic familiarity elicited a broadly distributed effect between 200 and 300 ms after stimulus onset, possibly representing the activation of face recognition units. Between 300 and 450 ms, frontal effects were observed for semantic familiarity and recognition-related familiarity, while perceptual fluency was associated with a centro-parietally focused effect. Thus, familiarity arising from the retrieval of semantic information and recognition-related familiarity depend at least partly on the same neuronal circuits, while these are dissociable from those mediating perceptual fluency.

Intermetamorphosis in a Patient With Alzheimer's Disease

Back to table of contents Previous article Next article LetterFull AccessIntermetamorphosis in a Patient With Alzheimer's DiseaseFrédéric Assal, M.D., and Mario F. Mendez, M.D., Ph.D., Frédéric AssalSearch for more papers by this author, M.D., and Mario F. MendezSearch for more papers by this author, M.D., Ph.D., VA Greater Los Angeles and UCLA Alzheimer's disease Center, Los Angeles, CA (f.a., m.f.m.); VA Greater Los Angeles Neurobehavior Unit and Department of Neurology and Psychiatry, UCLA, Los Angeles, CA (m.f.m.)Published Online:1 May 2003AboutSectionsView EPUB ToolsAdd to favoritesDownload CitationsTrack Citations ShareShare onFacebookTwitterLinked InEmail SIR: Intermetamorphosis is a misidentification syndrome in which familiar persons are believed to have exchanged identities.1 We report a patient with clinically probable Alzheimer's disease who developed intermetamorphosis for several members of his family. Analysis of this patient's disturbance suggests abnormal processing between the normal face recognition units and the person identity node in the inferior temporal lobe.2–4Case ReportA 76-year-old right-handed man had a 3-year history of progressive decline in memory and cognition. His medical history was remarkable only for prior ethanol abuse. The patient had undergone an evaluation for dementia, including neurological examination, laboratory tests, and neuroimaging. Magnetic resonance imaging was significant only for prominence of the temporal horns of the lateral ventricles, suggesting bilateral mesial temporal atrophy, and he was diagnosed with clinically probable Alzheimer's disease.About 2 years into his illness he began misidentifying family members. Initially, he misidentified his wife as his deceased mother and, later, as his living sister. His substitution of his sister for his wife became firmly established and could not be corrected with explanation. When confronted, he subsequently claimed that he had never been married or that his wife had left him. Gradually, the patient began misidentifying his son as his brother, his daughter as another sister, and his granddaughter as his daughter. These misidentifications persisted even on the telephone. Moreover, on different clinic visits he described the hospital as a meeting hall or a church where he visited to socialize or to help out.On examination he was cooperative, attentive, and euthymic. His Mini-Mental State Examination Score was 16 (out of 30), and his examination showed deficits in verbal fluency, confrontation naming, delayed recall, complex constructions, and abstractions. The elementary neurological examination was otherwise normal.We tested the patient with photographs of familiar persons from a photo album. Positive responses were counted when the photo subjects were positively identified or named. The control was his wife, who scored 100% correct answers. The results showed 100% misidentification of his wife as his sister in photographs with 100% persistence of the misidentification over the past without a retrograde gradient, and correct identification of himself through different ages. He did not have prosopagnosia (he recognized 80% of the photographs) and could recognize his own face in a mirror.CommentWe present a rarely described intermetamorphosis syndrome occurring in a patient with Alzheimer's disease. The patient did not have facial recognition difficulty, a temporal memory gradient for recalling familiar faces, or a failure to integrate prior memories of familiar faces with currently perceived faces. The misidentifications were not limited to the visual modality, as they occurred with the persons' voices as well. This points to a preservation of the abnormal “person identity nodes” across time and across modalities.These findings, as well as the reduplicative misidentification of the hospital, suggest an inability to correctly identify the source of his sense of familiarity for family members as a result of temporal lobe disease. Frontal lobe dysfunction also may contribute to his misattribution of identity.5References1 Courbon P, Tusques J: Illusions d'intermétamorphose et de charme. Ann Médico Psychol 1932; 90:401-405Google Scholar2 Ellis HD, Young AW: Accounting for delusional misidentifications. Br J Psychiatry 1990; 157:239-248Crossref, Medline, Google Scholar3 Breen N, Caine D, Coltheart M: Models of face recognition and delusional misidentification: a critical review. Cogn Neuropschychol 2000; 17:55-71Crossref, Medline, Google Scholar4 Hudson AJ, Grace GM: Misidentification syndromes related to face specific area in the fusiform gyrus. J Neurol Neurosurg Psychiatry 2000; 69:645-648Crossref, Medline, Google Scholar5 Tomita H, Ohbayashi M, Nakahara K, et al: Top-down signal from prefrontal cortex in executive control of memory retrieval. Nature 1999; 401:699-703Crossref, Medline, Google Scholar FiguresReferencesCited byDetailsCited byPsychiatry and Clinical Psychopharmacology, Vol. 28, No. 1Neuropsychologia, Vol. 45, No. 13 Volume 15Issue 2 May 2003Pages 246-247 Metrics History Published online 1 May 2003 Published in print 1 May 2003

Differential involvement of episodic and face representations in ERP repetition effects.

The purpose of this study was to disentangle the contribution of episodic-perceptual from pre-existing memory representations of faces to repetition effects. ERPs were recorded to first and second presentations of same and different photos of famous and unfamiliar faces, in an incidental task where occasional non-targets had to be detected. Repetition of same and different photos of famous faces resulted in an N400 amplitude decrement. No such N400 repetition-induced attenuation was observed for unfamiliar faces. In addition, repetition of same photos of faces, and not different ones, gave rise to an early ERP repetition effect (starting at approximately 350 ms) with an occipito-temporal scalp distribution. Together, these results suggest that repetition effects depend on two temporally and may be neuro-functionally distinct loci, episode-based representation and face recognition units stored in long-term memory.

Priming visual face-processing mechanisms: electrophysiological evidence.

Accumulated evidence from electrophysiology and neuroimaging suggests that face perception involves extrastriate visual mechanisms specialized in processing physiognomic features and building a perceptual representation that is categorically distinct and can be identified by face-recognition units. In the present experiment, we recorded event-related brain potentials in order to explore possible contextual influences on the activity of this perceptual mechanism. Subjects werefirst exposed to pairs of small shapes, which did not elicit any face-specific brain activity. The same stimuli, however, elicitedface-specific brain activity after subjects saw them embedded in schematic faces, which probably primed the subjects to interpret the shapes as schematic eyes. No face-specific activity was observed when objects rather than faces were used to form the context. We conclude that the activity of face-specific extrastriate perceptual mechanisms can be modulated by contextual constraints that determine the significance of the visual input.

Repetition priming for familiar and unfamiliar faces in a sex-judgment task: evidence for a common route for the processing of sex and identity.

Repetition priming for faces was examined in a sex-judgment task given at test. Priming was found for edited, hair-removed photos of unfamiliar and familiar faces after a single presentation at study. Priming was also observed for the edited photos when study and test faces were different exemplars. Priming was not observed, however, when sex judgments were made at test to photos of complete, hair-included faces. These findings were interpreted by assuming that, for edited faces, internal features are attended, thereby activating face-recognition units that support performance. With complete faces, however, participants provided speeded judgments based primarily on the hairstyle. It is suggested that, for both familiar and unfamiliar faces, a common locus exists for the processing of the identity of a face and its sex. A single face-recognition model for the processing of familiar and unfamiliar faces is advocated.

Names and Identification: An Access Problem

Abstract information about individuals unless given their names, which, however, he could not produce. It is proposed that CB's previously unreported pattern of symptoms results from multiple functional disconnections isolating the collection of identity-specific semantic information from: (1) the corresponding phonological form, hence proper name anomia; (2) face recognition units, preventing faces being named and matched to semantic information; (3) general semantics, preventing the elements of a definition referring to an individual eliciting further individual specific information. The case is discussed together with similar cases and relevant theories that recently appeared in the literature. The peculiar role of persons' proper names in the semantic system is highlighted. Unlike common names, which denote categories, proper names are thought to just label unique sets of casually clustered attributes constituting a person's identity-specific semantics.

Names and identification: An access problem

information about individuals unless given their names, which, however, he could not produce. It is proposed that CB's previously unreported pattern of symptoms results from multiple functional disconnections isolating the collection of identity-specific semantic information from: (1) the corresponding phonological form, hence proper name anomia; (2) face recognition units, preventing faces being named and matched to semantic information; (3) general semantics, preventing the elements of a definition referring to an individual eliciting further individual specific information. The case is discussed together with similar cases and relevant theories that recently appeared in the literature. The peculiar role of persons' proper names in the semantic system is highlighted. Unlike common names, which denote categories, proper names are thought to just label unique sets of casually clustered attributes constituting a person's identity-specific semantics.

Prosopagnosia and Alexia Without Object Agnosia

Following a trauma causing bilateral posterior brain damage, a patient complained of dyslexia and prosopagnosia, but not object agnosia. On testing she showed intact recognition of object drawings, even when it was assessed with perceptually demanding tasks such as Ghent's overlapping figures and Street completion test. This pattern of deficit is inconsistent with Farah's (1990) prediction that the simultaneous occurrence of alexia and prosopagnosia is invariably associated with object agnosia. The patient's reading performance had the features typically found in letter-by-letter readers. On face tests, she showed a discrepancy between the impairment exhibited in familiarity recognition and famous face naming and the correct (though slow) performance in matching the names of famous persons with their photographs. This apparent contradiction was clarified by showing that the patient had maintained the ability to generate the mental images of famous faces in response to the presentation of their names. We assume that face recognition units were intact, but partially disconnected from the output of perceptual processing.

False Recognition of Unfamiliar Faces Following Right Hemisphere Damage: Neuropsychological and Anatomical Observations

False recognition of unfamiliar faces was investigated in patients with focal right hemisphere damage (RHD) in order to define the neuropsychological and anatomical correlates of the recognition impairment and examine its relationship to prosopagnosia. Findings are discussed within the framework of the Bruce and Young (1986) model of face processing. Although false recognition and prosopagnosia were both present in some RHD patients, the two types of face recognition impairments were dissociable in others. Processing deficits in subjects with both false recognition and prosopagnosia were associated with posterior right hemisphere lesion sites and included severe face perception impairment and partial damage to face recognition units (FRUs). Prosopagnosia without false recognition was seen following near complete destruction of FRUs, but this type of dissociation could also occur when FRUs become disconnected. The opposite dissociation, false recognition without prosopagnosia, was observed following right prefrontal damage. We propose that false recognition in frontal patients results from the breakdown of strategic decision making and monitoring functions critical for determining whether a face is indeed that of a familiar person or whether there is merely a resemblance to a known individual. False recognition following prefrontal damage may also be related to confabulation, in which case familiarity or even specific identity are erroneously attributed to facial stimuli without the activation of an underlying memory representation.

Identification of famous faces and famous names in early Alzheimer's disease. Relationship to anterograde episodic and general semantic memory.

We assessed remote memory in 33 patients with dementia of Alzheimer's type (DAT) with Mini-Mental State Examination (MMSE) scores between 17 and 30, and 30 matched controls using a Famous Faces Test and Famous Names Test designed to assess face recognition, identification and naming, and name recognition and identification, respectively, together with a range of anterograde episodic and semantic memory tests. Patients with DAT were impaired on all components of the remote memory tests, i.e. famous face recognition, identification and naming, and famous name recognition and identification. There was also evidence of a modest temporal gradient, with relatively greater impairment of more recent memory, which may be artefactual resulting from the very insidious onset of their anterograde amnesia. In contrast to the uniform impairment of anterograde memory, there was considerable heterogeneity in performance on remote memory. Although the DAT patient group's performance on remote memory measures was impaired with respect to controls, some patients had significant impairment on all measures, whereas others had intact remote memory. Overall, there was only a weak correlation between dementia severity and remote memory, and no correlation between performance on the Faces and Names tests and measures of anterograde memory. At a cognitive level, the deficit in face and name processing in DAT involved recognition, identification and naming. This would suggest that so called 'face and name recognition units', semantic knowledge of famous persons and post-semantic processing are all affected by the disease. There was also supporting evidence for the concept that recognition of famous faces and names both draw on common sources. Similar results were found for face and name identification. This suggests that face and name recognition units are closely linked, and that identification of a face or name accesses the same central pool of semantic knowledge regarding the famous person. Performance on famous names tests correlated, to a limited degree, with that on general semantic tests, suggesting that knowledge of famous people, at least as accessed by names, is associated with general semantic memory. By contrast, no correlation was found between performance on the famous faces and on other general semantic tasks.