L5.2 faces

Face Recognition in Cognition and Perception

Overview of Lecture Content

  • Main Questions:
      - Are faces special?
      - What are the behavioral and neural markers associated with face processing?
      - How do face recognition models explain the differences between faces and objects?

Behavioral Markers of Face Recognition

  • Holistic processing tasks:
      - Understanding integration of facial features into a cohesive perceptual representation.

  • Tasks used to measure holistic processing:
      - Part-Whole Task (Tanaka & Farah, 1993):
        - Recognition of a face part is more accurate when presented within the whole face than in isolation.
      - Face Inversion Task (Yin, 1969):
        - Upright faces are recognized significantly better than inverted ones; effect is greater compared to objects.
      - Standard Composite Task (Hole, 1994; Young et al., 1987):
        - Difficulty in perceiving half a face when aligned with a misaligned complementary half to assess holistic processing.

Neural Markers of Face Recognition

  • Fusiform Face Area (FFA):
      - Plays a key role in face recognition and shows higher activation for faces than objects, though ~20% may show opposite response.

  • Occipital Face Area (OFA):
      - Involved in early stages of face processing.

  • N170:
      - An event-related potential that is larger for faces, indicating face-specific processing.

Perceptual Impairments relating to Face Recognition

  • Prosopagnosia:
      - “Face blindness,” impairment in recognizing familiar faces.
      - Acquired prosopagnosia: caused by brain damage.
      - Developmental prosopagnosia: occurs without evident brain damage.
      - Often co-occurs with impaired object recognition.

Differences Between Faces and Objects

  • Modularity Hypothesis vs. Expertise Hypothesis:
      - Modularity Hypothesis: Faces are processed via a dedicated cognitive module.
      - Expertise Hypothesis: Holistic processing is not unique to faces but applies to objects of expertise (e.g. cars, birds).

Face Recognition Models

  • Bruce and Young Model (1986):
      - Components:
        - Structural Encoding: Various representations of faces.
        - Expression Analysis: Inferring emotional states from features.
        - Facial Speech Analysis: Aiding speech perception from lip movements.
        - Face Recognition Units (FRU): Structural information about known faces.
        - Person Identity Nodes (PIN): Information about individuals (occupation, interests).
        - Name Generation: Process of recalling a person's name.
      - Evaluation:
        - Explains face recognition's uniqueness, modularity, and various impairments but fails to address dynamic and social aspects.

  • Interactive Activation and Competition Model (IAC; Burton et al., 1990):
      - Explains context and variability in name recall, yet is abstract with no neural pathways considered.

  • Distributed Human Neural System Model (Haxby et al., 2000):
      - Divided into core (OFA, FFA, pSTS-FA) and extended systems for processing invariant (identity) and changeable aspects (gaze, expression).
      - Evaluation highlights the need for dynamic processing capabilities and interactions with other cognitive functions.

  • Face Space Model (Valentine, 1991):
      - Spatial representation of faces, with average faces at the center and distinctive faces toward the periphery.
      - Evaluation notes mechanisms and social/emotional information are poorly addressed.

Applied Importance of Face Recognition Models

  • Evidence in real-world applications:
      - Eyewitness Testimony: Reliability issues based on face recognition capabilities.
      - Clinical Disorders: Understanding and diagnosing prosopagnosia.
      - Forensic and Security Applications: Utilizing face recognition in security systems.
      - Everyday Memory: Implications for memory retrieval and social interactions.
      - Social and Emotional Processing: Understanding interpersonal recognition and interactions.

Summary

  • Key Takeaways: Faces require unique cognitive processing abilities which intersect between behavioral performance (holistic processing) and neurological mechanisms (FFA, OFA, N170).

  • Discussion on whether these processes arise from innate modules or through expertise shows a rich area of ongoing research and practical implications for various fields.

References

  • Kietzmann, T. C., Poltoratski, S., König, P., Blake, R., Tong, F., & Ling, S. (2015). The occipital face area is causally involved in facial viewpoint perception. Journal of Neuroscience, 35(50), 16398-16403.

  • Key literature referenced throughout:
      - Tanaka & Farah (1993), Yin (1969), Hole (1994), Young et al. (1987), Bruce & Young (1986), Burton et al. (1990), Haxby et al. (2000), Valentine (1991).

  • Studies on prosopagnosia: Busigny et al. (2010), Germine et al. (2010), Geskin & Behrmann (2018).

Conclusion

  • Understanding face recognition encompasses various interdisciplinary approaches uniting cognitive psychology, neuroscience, and applied practices. The ongoing inquiry into face processing continues shaping these fields' theoretical frameworks and practical applications.