Background

1
Can context affect gender processing?
ERP Evidence about differences between lexical
and stereotypical gender H. Kreiner,
S. Mohr, K. Kessler, and Garrod, S. CCNI,
University of Glasgow, UK hamutal_at_psy.gla.ac.uk

• The question
• Readers experience processing difficulty when an
  anaphor (herself) refers to an antecedent
  (minister) that mismatches in gender (e.g., (a),
  (b)).
• Yesterday the king left London after reminding
  herself about the letter.
• After reminding herself about the letter, the
  king immediately went to the meeting.
• This difficulty evidenced in both eye-tracking1
  and ERP studies2 and has been attributed to a
  gender clash between the pronoun and the
  antecedent.
• In this paper we ask whether, and in which
  conditions,
• discourse context can modulate this clash and how
  this is reflected in EEG parameters.

• Background
• Dissociation
• Lexical-gender (king) - recovered directly from
  the lexicon
• Stereotypical-gender (minister) inferred from
  pragmatic information3
• Hence Stereotypical gender more sensitive to
  context effects4.
• Support from eye-tracking studies 1
• In anaphora (a) both noun types lead to similar
  mismatch-effect.
• In cataphora (b) mismatch-effect shown only for
  lexical-gender nouns
• Conclusion Stereotypical, unlike lexical
  gender, can be overridden when gender is
  prespecified by context (b).
• Inconsistency with ERP findings
• Lexical features can be overridden by context5.
• P600 mismatching-effects for both stereotypical
  lexical nouns
• No qualitative difference in the processing of
  these noun types2.

Method Participants. 20 native English speakers
in each experiment. Materials. 160 ANAPHORA
sentences (e.g. a, b) in Experiment 1 (a1)
Yesterday the king/minister left London after
reminding himself about the letter. (a2)
Yesterday the king/minister left London after
reminding herself about the letter.
160 CATAPHORA sentences (e.g., c, d) in
Experiment 2 (b1) After reminding himself
about the letter, the king/minister immediately
went to the meeting. (b2) After reminding
herself about the letter, the king/minister
immediately went to the meeting. Design. 2X2 -
Gender Type (lexical/stereotypical) X Matching
(match/mismatch). Procedure. ? Silent reading,
word by word visual presentation ?
50 fillers 25 Comprehension questions. Analysi
s. ? 200 msec. epoch pre-stimulus onset was used
as reference. ? EEG time-locked to
onset of target (pronoun in Exp.1 role-noun in
Exp.2 ? Data filtering, Automatic
artifact correction rejection using BESA 5.1.6.
Anaphora (Exp.1)
Cataphora (Exp.2)



250-500 ms. from target onset
Lexical
Stereotypical
500-750 ms. from target onset
Lexical
Stereotypical

• Findings (500-750 msec. from target onset)
• ERP analysis mismatching effect for both
  stereotypical and lexical role nouns in anaphora
  and cataphora.
• ERP-components no qualitative difference
  however amplitudes - modulated by sentence
  structure and noun-type
• Anaphora P600-like mismatching effect - larger
  for lexical compared to stereotypical gender
• Cataphora sentences only subtle interaction
  between gender matching and noun type

• Discussion
• Why does the seemingly pragmatic effect of
  stereotypical gender mismatching elicits a
  P600-like component and not an N400?
• Sentence structure modulates ERP wave form
  what does this reflect?
• Differences in discourse alignment (the
  antecedent precedes the reference in anaphora and
  vice versa in cataphora)?
• Different types of target words (pronouns in
  anaphora, nouns in cataphora) elicit different
  processes?
• What can we learn from different EEG analyses?
• Time-Frequency Representations Gamma mismatching
  effects diverge between anaphora and cataphora.

Match
Mismatch
Match
Mismatch
0-1000 ms. from target onset
Lexical
References 1 Pollatsek, Bolozky, Well,
Rayner, 1981 2 Deutsch Rayner, 1999 3
Farid Grainger, 1996 4 Engbert Kliegl,
2004 5 MacDonald, Mac Cumhaill, Tamariz
Shillcock (in preparation).
Stereotypical
2
References1 Kreiner, H., Sturt, P. Garrod,
S. (2008). Processing definitional and
stereotypical gender in reference resolution
Evidence from eye- movements. Journal of Memory
and Language, 58, 239261. 2 Osterhout, L.,
Bersick, M., McLaughlin, J. (1997). Brain
potentials reflect violations of gender
stereotypes. Memory Cognition, 25, 273285.
3 McKoon, G. and Ratcliff, R. (1992).
Inference during reading. Psychological Review,
99, 440-466.4 Carreiras, M., Garnham, A.,
Oakhill, J. and Cain K. (1996). The use of
stereotypical gender information in constructing
a mental model Evidence from English and
Spanish. The quarterly Journal of Experimental
Psychology, 49A(3),639-663.5 Nieuwland, M. S.,
van Berkum, J. J. A. (2006). When peanuts fall
in love N400 evidence for the power of
discourse. Journal of Cognitive Neuroscience,
18(7), 10981111.