Face and object encoding under perceptual load: ERP evidence

Markus F. Neumann*, Tarik N. Mohamed, Stefan R. Schweinberger

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

57 Scopus citations


According to the perceptual load theory, processing of a task-irrelevant distractor is abolished when attentional resources are fully consumed by task-relevant material. As an exception, however, famous faces have been shown to elicit repetition modulations in event-related potentials - an N250r - despite high load at initial presentation, suggesting preserved face-encoding. Here, we recorded N250r repetition modulations by unfamiliar faces, hands, and houses, and tested face specificity of preserved encoding under high load. In an immediate (S1-S2) repetition priming paradigm, participants performed a letter identification task on S1 by indicating whether an "X" vs. "N" was among 6 different (high load condition) or 6 identical (low load condition) letters. Letter strings were superimposed on distractor faces, hands, or houses. Subsequent S2 probes were either identical repetitions of S1 distractors, non-repeated exemplars from the same category, or infrequent butterflies, to which participants responded. Independent of attentional load at S1, an occipito-temporal N250r was found for unfamiliar faces. In contrast, no repetition-related neural modulation emerged for houses or hands. This strongly suggests that a putative face-selective attention module supports encoding under high load, and that similar mechanisms are unavailable for other natural or artificial objects.

Original languageEnglish
Pages (from-to)3021-3027
Number of pages7
Issue number4
StatePublished - 14 Feb 2011
Externally publishedYes


  • Attention
  • ERP
  • N170
  • N250r
  • Perceptual Load Theory
  • Repetition

ASJC Scopus subject areas

  • Neurology
  • Cognitive Neuroscience


Dive into the research topics of 'Face and object encoding under perceptual load: ERP evidence'. Together they form a unique fingerprint.

Cite this