Introspection - Criticisms

Criticisms

Already in the 18th century authors had criticized the use of introspection, both for knowing one's own mind and as a method for psychology. David Hume pointed out that introspecting a mental state tends to alter the very state itself; a German author, Christian Gottfried Schütz, noted that introspection is often described as mere "inner sensation", but actually requires also attention, that introspection does not get at unconscious mental states, and that it cannot be used naively - one needs to know what to look for. Immanuel Kant added that, if they are understood too narrowly, introspective experiments are impossible. Introspection delivers, at best, hints about what goes on in the mind; it does not suffice to justify knowledge claims about the mind. Similarly, the idea continued to be discussed between John Stuart Mill and August Comte. Recent psychological research on cognition and attribution has asked people to report on their mental processes, for instance to say why they made a particular choice or how they arrived at a judgment. In some situations, these reports are clearly confabulated. For example, people justify choices they have not in fact made. Such results undermine the idea that those verbal reports are based on direct introspective access to mental content. Instead, judgements about one's own mind seem to be inferences from overt behavior, similar to judgements made about another person. However, it is hard to assess whether these results only apply to unusual experimental situations, or if they reveal something about everyday introspection. The theory of the adaptive unconscious suggests that a very large proportion of mental processes, even "high-level" processes like goal-setting and decision-making, are inaccessible to introspection. Indeed, it is questionable how confident researchers can be in their own introspections.

One of the central implications of dissociations between consciousness and meta-consciousness is that individuals, presumably including researchers, can misrepresent their experiences to themselves. Jack and Roepstorff assert, ‘…there is also a sense in which subjects simply cannot be wrong about their own experiential states.’ Presumably they arrived at this conclusion by drawing on the seemingly self-evident quality of their own introspections, and assumed that it must equally apply to others. However, when we consider research on the topic, this conclusion seems less self-evident. If, for example, extensive introspection can cause people to make decisions that they later regret, then one very reasonable possibility is that the introspection caused them to ‘lose touch with their feelings’. In short, empirical studies suggest that people can fail to appraise adequately (i.e. are wrong about) their own experiential states.

Another question in regards to the veracious accountability of introspection is if researchers lack the confidence in their own introspections and those of their participants, then how can it gain legitimacy? Three strategies are accountable: identifying behaviors that establish credibility, finding common ground that enables mutual understanding, and developing a trust that allows one to know when to give the benefit of the doubt. That is to say, that words are only meaningful if validated by one's actions; When people report strategies, feelings or beliefs, their behaviors must correspond with these statements if they are to be believed.

Even when their introspections are uninformative, people still give confident descriptions of their mental processes, being "unaware of their unawareness". This phenomenon has been termed the introspection illusion and has been used to explain some cognitive biases and belief in some paranormal phenomena. When making judgements about themselves, subjects treat their own introspections as reliable, whereas they judge other people based on their behavior. This can lead to illusions of superiority. For example, people generally see themselves as less conformist than others, and this seems to be because they do not introspect any urge to conform. Another reliable finding is that people generally see themselves as less biased than everyone else, because they are not likely to introspect any biased thought processes. These introspections are misleading, however, because biases work sub-consciously.

One experiment tried to give their subjects access to others' introspections. They made audio recordings of subjects who had been told to say whatever came into their heads as they answered a question about their own bias. Although subjects persuaded themselves they were unlikely to be biased, their introspective reports did not sway the assessments of observers. When subjects were explicitly told to avoid relying on introspection, their assessments of their own bias became more realistic.