Inference Processes

1
Inference Processes

• Inference another way we go beyond (elaborate
  on, use) information. Inferring other
  properties, characteristics of target.
• Where do inferences come from?
• Cognitive structures the knowledge, beliefs
  and expectancies stored in memory, relevant to
  some domain.
• Cognitive structures guide inference processes,
  just like every aspect of information processing.
  

2
Inference Processes

• Early examples of inference in social perception
  research
• Implicit Personality Theories. First approach
  to studying cognitive structures in person
  perception. If person has Trait X, how likely
  also has Trait Y?
• Asch research on forming first impressions.
  Ss completed checklist indicating other traits
  describing target. These are inferences.

3
Inference Processes

• Correspondent Inference Theory (Jones Davis,
  1965)
• Dispositional From Acts to Dispositions The
  Attribution Process in Person Perception.
• (incl. traits, attitudes, emotions, motives)
• What is correspondent inference?
• Why make correspondent inferences?
• When make correspondent inferences?
• How make correspondent inferences?

4
Inference Processes

• When make correspondent inferences?
• D
• B
• S
• - normativeness
• - desirability
• - choice

5
Inference Processes

• Test of correspondent inference theory (Jones
  Harris, 1967)
• Ps read essay written by target person
• Pro or con position on issue
• Choice vs. no choice
• Question what is authors true attitude on
  issue?

6
Inference Processes

• Jones Harris (1967) results
• Essay Direction
• Condition Pro-Castro Anti-Castro
• Choice 59.62 17.38
• No Choice 44.10 22.87
• Correspondent inference occurs even when it
  shouldnt
• Apparently oblivious to situational influences
• Robust finding
• Correspondence bias

7
Inference Processes

• oblivious to situational influences. -- what
  does that mean? Several possibilities
• May believe situation has little impact on
  behavior
• May not be aware of situational constraints on
  behavior
• May underestimate power of situational factors
• May fail to sufficiently adjust or correct
  initial inferences

8
Inference Processes

• Correspondence bias a pervasive tendency
• Questions about process
• Spontaneous trait inferences (STIs)
• STI -- based on encoding specificity (Rdng. 18,
  Winter Uleman, 1984)

9
Inference Processes

• STI false recognition paradigm (Todorov
  Uleman, 2002, 2003, 2004)
• Study phase photos behavior-descriptive
  sentences
• Test phase photos traits was trait word
  included in sentence?
• Traits implied by behavior
• Control traits
• DV
• Recognition accuracy
• Latency

10
Inference Processes

• Correspondent inferences/STIs fixed or
  changeable?

11
Inference Processes

• Gilbert, Pelham, Krull (1988 Rdng. 19)
• Three stages in inference process
• 1) Categorization identify behavior
• 2) Characterization trait inference
• 3) Correction situational correction revision
  of trait inference in light of context
  information
• Categorization and characterization are
  relatively automatic
• Correction is a more controlled process that
  requires cognitive resources

12
Inference Processes

• Gilbert et al. -- Method
• Video of an woman acting nervously (no sound)
• IV Supposedly discussing one of two topics
• Anxiety provoking topic
• Neutral topic
• IV Participants watched with cognitive load or
  not
• Rated how anxious the woman appeared

13
Inference Processes

• Gilbert et al. Results
• Discussion Topic One Task Two Tasks
• Relaxing 10.31 9.28
• Anxious 7.79 8.88

14
Inference Processes

• Correction requires motivation and capacity
• Motivation and capacity are not required to make
  initial dispositional inferences
• Default/automatic process ? Dispositional
  Inferences
• Controlled correction ? Take situation into
  account

15
Inference Processes

• Self and inference processes
• Earlier in term, talked about
• Categories, categorization (article by Medin)
• Social identity theory starts with
  categorization into ingroup and outgroup

16
Inference Processes

• Another theory in the social identity tradition
  self-categorization theory.
• Categorize into ingroup and outgroup -- involves
  categorizing self
• One outcome from categorization perceiving all
  category members alike, having same attributes
• Self-categorization ? perceived similarity of
  self to other group members, having same
  attributes
• Self-stereotyping accepting group-stereotypic
  attributes as characteristic of self
• Thus, inferences about self based on group
  membership.

17
Inference Processes

• Person Perception and inference processes
• Already well covered virtually all studies
  weve talked about have examined inferences in
  perceiving others

18
Inference Processes

• Stereotypes and inference processes
• Stereotyping is inferencing inferences about
  groups
• Stereotype cognitive structure containing ones
  knowledge, beliefs, and expectancies about some
  group
• Stereotype activation ? inferences about
  characteristics of group and its members
• Hence, stereotyping and inferences are intimately
  linked

19
Inference Processes

• Stereotypes and spontaneous trait inferences
  (Wigboldus, Dijksterhuis, van Knippenberg,
  2003)
• We know -- people make STIs near-automatic.
• Stereotype activation
• increases accessibility of stereotype-consistent
  traits
• decreases accessibility of stereotype-inconsistent
  traits
• How do stereotypes influence STIs? Rationale
  consider relation between stereotype and trait
  implied by behavior.
• Consistent traits STIs should occur easily
  (Professor won science quiz.) Smart?
• Inconsistent traits STIs should not occur
  (Garbage man won science quiz.) Smart?

20
Inference Processes

• Method recognition probe paradigm
• Sentences presented on screen
• Followed by probe word
• Task indicate if word was in sentence (correct
  answer No) RT measured.
• On some trials, probe words were traits implied
  by behavior (e.g., smart). If STI made, should
  slow decision process (RT).
• Trait implied by behavior is therefore consistent
  (professor) or inconsistent (garbage man) with
  stereotype

21
Inference Processes

• Results
• Stereotype Consistency
• consistent inconsistent
• Mean RT 913 873
• Conclusion takes more time to say No when
  implied trait consistent with stereotype than
  when inconsistent with stereotype.
• Why? Did people infer trait (smart) from
  sentence, or did activation of stereotype lead to
  trait? Either process would make trait
  accessible and make No response harder (gtRT).

22
Inference Processes

• Next study include neutral behaviors (e.g.,
  Professor/garbage man cycles through streets),
  followed by probe (smart).
• If results due to stereotype activation, results
  for neutral sentences should be same as for
  consistent sentences. Why?
• If results due to STIs, then results for neutral
  and consistent sentences should differ. Why?

23
Inference Processes

• Results
• Stereotype Consistency
• consis neutral inconsis
• Mean RT 862 803 834
• Conclusion longer RT for consistent sentences
  because both stereotype behavior imply same
  trait
• Not so for inconsistent sentences
• Ex. stereotype of garbage man blocks STI that
  winning science quiz means person is smart,
  hence quicker RT to say No than for consistent
  sentences.