Nisbett and Wilson 1997 - Educated guess work
- introspection educated guess work
- inaccurate insights may still affect behaviour
- people report cognitive processes inaccurately
- based on implicit causal theories
- little/ no direct access to higher order cognitive processes
James 1890
pg.185
we discover states of consciousness through introspective observation
Lashey 1923
introspection may make the preliminary survey but it must be followed by the chain of transit objective measurement
Comte’s paradox 1842
thinking subject cannot divide himself into two parts, one reasoning and one observing
- introspection is limited as an experimental method as it is impossible to be objective
Dressing Gown Study - Nissbett and Wilson 1977
-Passersby were asked to evaluate what they were told were four different nightgowns and four different pairs of stockings. In fact, all four items were identical.
ppts asked to choose their preference of nightgown out of 4 different ones and justify their choice
40% chose option 4 (most right)
> people argued their choice was made due to quality > nightgowns were identical
> position never stated as reason for choice and when probed people denied
> suggests spatial location impacts peoples choices
Petitmenging 2013 - Choice Blindness and Improving Introspection
–incorporated elicitation interview > when interview occurred, ppts detected the manipulation 80% of the time compared to 33% of the time in the classic version
showed that subjects can be interactively guided to become better aware of their experience thereby overturning the choice blindness results
- first experimentally verified evidence for direct efficacy of a second-person approach to measure of consciousness
-fact that we are normally not skilled at accessing our experience does not mean that this experience is inaccessible in principle
Kruger & Dunning 1999 - Unskilled and Unaware
- argued unskilled people’s incompetence robs their meta cognitive ability to realise their deficits
-found that in a range of different tests , 4 studies, (including logical reasoning and grammar, humour) participants with the poorest performance (bottom quartile) misjudge their incompetence the most - after training and improving skills , paradoxically both cognitive and meta cognitive performance was improved and ppts became more aware of their incompetence
Petitmengin (2006) - Elicitation Interview
arguing that the subject’s first-person introspection should be supported by an interactive second-person approach
suitably trained interviewer helps the subject to come into contact with their experience and to describe it
> interviewer must be well trained and training must be standardised to avoid bias
- checked for internal consistency (Hurlburt & Heavey,2002), for correlations with other standard psychological measures (Heavey & Hurlburt, 2008), and for correlations with neuroscientific data (Petitmengin, Navarro, & Le Van Quyen, 2007)
Introspection can reveal…
-better insight into metacognitive processes
- wider cognitive processes
- perceptual processes
Methods used to evaluate introspection
- verbal report
- choice blindness
- deception based studies
Nisbett and Wilson - Movie study
- ppts watched a movie and rated enjoyment levels (MC)
- one condition distractors > despite this ratings of enjoyment consistent across groups indicating not effect of distractors
- however 55% of ppts reported distractors influenced rating
- ppts made incorrect introspections on whether factors (distractors) affected judgement of enjoyment of movie
Nisbett and Wilson - Movie study (CRITICISM)
- difficult for ppts to compare to how they would judge their enjoyment without distractions as it is theoretical scenario
- could lead to inference rather than introspection
- difficult to eliminate results of individual differences (some may be more easily distracted than others) > however it’s not clear how this might artifactually create the effects observed, since this variability would add noise (rather than bias) to the results.
de Camp, Nisbett and Wilson - No true introspection
argument that there is no true introspection and rather reports are based on a priori, implicit causal theories
Johansson 2008 - Choice Blindness
- ppts given choice of two faces asked to rate one as more attractive > ppts given card face down
- ppts asked to explain their choice
3/15 trials > ppts given photograph they did not choose
Johannson 2008 - Choice Blindness (FINDINGS)
- 75% of participants failed to notice the switch, showing choice blindness.
- 1/5 of all manipulations detected
- Participants provided detailed explanations for their choice, even though they actually rejected the chosen picture.
- can be applied to taste and smell, consumer choices, reasoning problems, moral and political issues
- quality of verbal reports similar for true and false introspections
Johannson - 2008 - Choice Blindness - Post Hoc Constructions
- In a post-test interview, 84% of participants claimed they would notice if their choices were switched, but many expressed surprise and disbelief when the truth was revealed.
- Participants remained confident in their explanations and showed similar emotional responses for chosen and non-chosen pictures.
- Choice blindness had an impact on future preferences, with participants sometimes coming to prefer the previously rejected alternative based on their explanations and feedback.
Maier 1931 - Two string problem
Experiment involved cords hanging from the ceiling in a room with objects.
Goal was to tie the two cords together, which were placed apart.
Participants solved the problem using a pendulum method within 45 seconds.
Only 1/3 of participants acknowledged the cue’s help.
In another condition, a useless cue was given prior to the useful one.
Subjects reported the useless cue as helpful and denied the usefulness of the useful cue.
Froese 2013 - Double Blind Interview
experimenter guides participants through introspection but the stimulus is novel also to the experimenter to avoid bias in their support
Ericson and Simon 1980 - Verbal Reports
- argue that criteria need to met to make it more likely that a conscious report is reliable
- must be concurrent
- must be attended in short term memory
- report must be direct rather than based on inference or a translation process (e.g. mental image to verbal description)
Valins & Ray 1967 - Attribution Paradigms
- Snake phobic participants watched slides whilst receiving occasional shocks
- Participants were wired for what they thought were heart rate recordings but were just rhythmic sounds.
- Slides contained series of snakes with the occasional word ‘shock’ on some slides
Valins & Ray 1967 - Attribution Paradigms (FINDINGS)
FINDINGS
- ‘shock’ slides accompanied with increase in heart rate whilst snake slides led to no effect
NEXT
- ppts asked to approach a 30inch boa constrictor
- Participants in the fake heart rate condition approached a boa constrictor more closely > theorised this was due to fake heart rate creating a new attribution for snake fearlessness as ppts assumed they were frightened by shock but not snake
In CONTRAST
- Participants aware of the extraneous sounds did not approach the snake as closely.
suggests manipulations of introspections can change behavior > Implications for therapy and potential new desensitization methods.
Generalizability of the task is a concern.
