|Morten Overgaard (2008), Scholarpedia, 3(5):4953.||doi:10.4249/scholarpedia.4953||revision #91387 [link to/cite this article]|
Introspection refers to an observation and, sometimes, a description of the contents of one’s own consciousness. Introspection is believed to be a reflexive, metacognitive process, attending to or thinking about oneself or what is currently being experienced by oneself. It has been much debated whether introspection is a valid and reliable scientific method.
Introspection and scientific methodology
The renewed interest in consciousness within experimental disciplines, especially cognitive neuroscience, has led to certain methodological challenges. Cognitive neuroscientists generally believe that objective data is the only reliable kind of evidence, and they will tend to consider subjective reports as secondary or to disregard them completely. For conscious mental events, this approach seems however futile: Subjective consciousness cannot be observed ‘from the outside’ with traditional objective means. Accordingly, some will argue, we are left with the challenge to make use of subjective reports within the framework of experimental psychology.
Some scientists have argued that there are other possible solutions. For instance, Persaud, McLeod & Cowey (2007) have argued that wagering in experimental situations, literally betting on the correctness of one’s own response to a stimulus, is an objective way of measuring how conscious a subject was of that stimulus. How well such methods actually do measure consciousness rather than other cognitive or emotional processes is however not yet investigated.
Fathers of introspective methodology
There are historical disagreements as to what constitutes a subject’s access to his or her own mind. Franz Brentano (1874) argued that a paradox exists in the relation between observations of ‘‘inner’’ mental states and “outer” objects. In order to observe and know about, say, an experience of a red apple, one must turn one’s attention from that outer object which was cause to the sensation. This should logically make the relevant experience cease to exist, thus also the attempted introspection. In other words, Brantano argued that any active introspection would immediately eliminate itself. The only possible kind of introspection, Brentano argued, is a passive inner perception consisting of a change of focus. Wilhelm Wundt (1907) adopted this interpretation of subjective knowledge but argued that the passive inner perception could become a scientific enterprise by systematically training subjects to report what thus is passively perceived. Wundt believed that one should only do experiments when one has external control over stimuli, as in perception, and that training is necessary to give reliable reports. Wundt’s introspective experiments were thus carried out under in principle replicable circumstances, with controlled stimuli that were claimed to be ‘‘passively observed with inner perception’’ by well-trained subjects. Although Wundt in many accounts is described as “father of introspective methodology”, a clear difference between a systematic use of ‘‘normal, non-introspective reports’’ and ‘‘introspective reports’’ is hard to find in his works (Lyons, 1986).
Historical critique of introspective methods
Comte raised two objections to a science based on introspection (Lyons, 1986). Comte’s first objection was that one cannot have an identity between the observer and the object of observation in science. He argued that the observer cannot be ‘‘split in two’’ so that one part observes the other, and, thus, observation of one’s own inner experiences is an impossible project. Comte’s second objection was of a more empirical kind: He argued that even if we would set aside the principal problems with introspection, it will generate unreliable and conflicting data. This second objection that seemed more of a claim than a real argument related to the controversies over data that came out of the laboratories in Cornell or Würzburg. Such controversies and the lack of success replicating results were likely what led to the increase of interest in behaviourism rather than any theoretical problem related to introspection (Overgaard, 2006).
Is introspection really retrospection?
By the term introspection, William James meant a kind of active observation. James was aware of the objections raised by Comte, and responded to them in a defence of introspective methodology. First of all, James argued that Comte cannot deny that we know about our own mental states, so when we cannot “split into two”, our best knowledge of our ‘‘inner states’’ are by way of memory: Attending to experiences we previously had. In seeing introspection as ‘‘retrospection’’, James also responded to Wundt’s and Brentano’s worry that active observation of mental states may change or destroy the experienced content.
One may speculate that one could have taken a different perspective in a response to Comte than the one James chose. Comte argued against a split of consciousness in ‘‘two parts’’ but stated no reason other than the ‘‘prima facie oddness’’ hereof. Today, some researchers have in different contexts argued exactly for the possibility that the self is not always acting as one, indivisible unit. Empirical support for such arguments has however mostly been more ‘‘extreme’’ cases, such as anosognosia for hemiplegia – a condition where patients insist they have preserved bodily functions even though they are paralyzed (Marcel, Tegner, & Nimmo-Smith, 2004).
One could also have raised the contrary argument that the description of introspection giving rise to a split in consciousness is in fact a misleading description. ‘‘The self” or the subject is obviously not identical to the content of his or her consciousness; for instance, the subject enjoys an uncountable number of conscious ‘‘states’’ throughout his or her lifetime. Were the subject identical to conscious content, he or she would be as many selves as possible number of contents, continuously beginning and ceasing to exist. Thus, introspection could be conceived of as a simple split between the subject observing his or her conscious content, which should serve no theoretical problem.
One problem arising with James’ solution, turning introspection into retrospection is this: if his solution is to work, the memory being actively observed must be an unconscious memory. If the memory was in fact conscious, introspection of ongoing mental events would be a possibility. Were the memory part of the subject’s consciousness, Comte’s objection would seemingly still apply. So in order to argue that introspection is retrospection, one is forced not to accept introspection of currently conscious states. However, in practice, would not an attending to the unconscious memory make its content conscious? Whereas such difficulties make James’ solution look less attractive, the problems contained in Comte’s second objection caused the most serious difficulties to the use of introspection in early experimental psychology.
Some authors (e.g. Schooler, 2002) seem to use the term introspection differently, to mean something more like ‘rationalization’. Among these authors, Nisbett and Wilson (1977) have proposed the still widely accepted view that introspective report have no place in scientific research. It could be argued. This conclusion, however, does not follow directly from the empirical evidence, they report. Subjects giving an introspective report about, say, whether they like a certain object, they may be giving a perfectly good and scientifically usable report of what they experienced even though they may be unable to explain the causal history behind liking the object. Nisbett and Wilson correctly rejected introspection as a methodology to learn about (some aspects of) choice and decision making, as their behavioural data suggested a very different explanation from the one that subjects themselves reported.
Another interpretation of the results could be, however, that in some unknown (but probably vast) number of situations, people do not have introspective access to their own cognitive processes. However, not surprisingly, they still have some experience and interpretation of their own actions. Thus, a conflict in data between subjective report and behavior could be interpreted to show that the subject’s experience differs from what can be analyzed from his or her behaviour, and, thus, it does not automatically follow that the introspective report is invalid.
It could be argued that even the most methodologically rigid experiments in cognitive science need introspective methodology at some level. For one thing, cognitive science often uses subjective reports. In experiments about consciousness, subjects are asked how certain they feel to give the correct report or which colour they have perceived without pre-existing methodological discussions how to handle subjective data empirically. Such reports are clearly introspective reports as they are specifically about conscious states.
But even if one stays completely clear of using verbal reports of any kind, introspective or non-introspective, there will always be some motivation for conducting an experiment in a certain way. For instance, a scientist may raise questions about the difference between seeing different colours, being in different emotional states, whether certain cognitive processes exist unconsciously, etc. The underlying motivation in all those instances must in the end be the scientist’s experience with his or her own conscious states, or, so to say, be based on introspection. Were it not the case that the scientist had introspective access to perceptions, emotions etc, he or she would not get the idea to raise scientific questions about them That is, even experiments that only make use of non-introspective evidence will, in psychology at least, rely on introspective evidence. This fact, it seems, makes it seem unthinkable to do experimental psychology, or perhaps any kind of psychology without introspection.
Introspection and consciousness
Today, many believe that introspection can function as one method among many other methods (analysis of behaviour, etc.) and that it is not intended to acquire knowledge about all aspects of human cognition. Some would even say that introspection is the key method to study consciousness (Jack & Roepstorff, 2003). It seems we are stuck with introspection if we want to know about consciousness. So, rather than discussing how to get rid of introspection, many recent papers argue that we should make use of introspection in scientific experiments in a more disciplined manner (Jack & Roepstorff, 2003, 2004; Overgaard, 2006).
Introspection and phenomenology
There are different interpretations of whether introspection should be seen as identical to or different from phenomenology (Overgaard, 2006). Most philosophers, however, seem to consider introspection as a sort of higher-order cognitive act, attending to “inner events”, whereas phenomenology is world-directed or “experience of being in the world”.
Philosophers and scientists devoted to phenomenology consider consciousness irreducible and it is taken as the starting point of any kind of analysis or statement. The fundamental approach, inspired by the work of Husserl, is a bracketing of ontological questions regarding the outside world, so that all every-day beliefs about what is real and what is not are methodologically suspended. From that point, one goes forward and carefully describes how the world appears to the observing subject (e.g. Varela, 1996; Varela & Shear, 1999).
Although there are a number of variations of phenomenology (some of which are primarily inspired by Eastern contemplative traditions), one can generally distinguish between a transcendental phenomenology and a phenomenological psychology. A transcendental phenomenology is the attempt to make claims about the nature of consciousness and objects in the world via a systematic descriptions of experiences. A phenomenological psychology is basically a scientific enterprise based upon a vocabulary of words that refer to conscious experiences (Overgaard, 2004). Although one may point to important ontological differences between transcendental phenomenology and the assumptions behind most versions of introspectionism, they are methodologically quite similar (Overgaard, Gallagher & Ramsøy, 2008).
The practice of introspection
In fact, very few recent articles have anything to say with regards to the actual practice of introspection. However, a few “guidelines” to such practice can be extrapolated from the theoretical (and, in few cases, experimental) literature:
- Insights obtained from introspection, can guide experimental design directly (Gallagher, 2003). That is, distinctions known to us by way of introspection can form questions for experimental science to answer.
- Subjects should be trained to “observe experiences” without prejudice. That is, experimental subjects may have conceptions about the nature of different kinds of conscious experience prior to an experiment, and subjects should therefore meet experimental situations in a “non-theoretical manner”. (Varela, 1996).
- The character of mental states should not be predetermined by the experimental scientist investigating them. Therefore, the scientist should therefore discuss the methodology for subjective reporting prior to the actual experiment (Ramsøy & Overgaard, 2004).
- Post hoc interviews should be carried out for retrospective examination of the subjects’ experiences during the experiment to gather information too rich or complicated to investigate during the actual experiment (Jack & Roepstorff, 2002). Such interviews can inspire a re-grouping of data.
The actual application of these methods, or others deriving from introspective or phenomenological approaches may inform and assist cognitive scientists to acquire much more exact data about what their subjects actually experience. In different areas of clinical work, better methods to obtain subjective reports may also be of help. Petitmengin, Navarro & Le Van Quyen (2007) have for instance applied a “neuro-phenomenological” approach to discover and prevent epileptic seizures.
- Brentano, F. (1874): Psychology from an empirical standpoint, Routledge & Kegan Paul
- Gallagher, S. 2003. Phenomenology and experimental design. Journal of Consciousness Studies 10 (9-10): 85-99.
- Jack, A. & Roepstorff, A. (2002): Retrospection and cognitive brain mapping: from stimulus-response to script-report, Trends in Cognitive Science, 6 (8), 333-339
- Jack, A. I., & Roepstorff, A. (2003). Trusting the Subject I, special issue of. Journal of Consciousness Studies, 10, 9–10.
- Jack, A. I., & Roepstorff, A. (2004). Trusting the Subject II, special issue of. Journal of Consciousness Studies, 11, 7–8.
- Lyons, W. (1986): The Dissapearance of Introspection, MIT Press
- Marcel, A. J., Tegnér, R., & Nimmo-Smith, I. (2004). Anosognosia for plegia: Specificity, extension, partiality and disunity of bodily unawareness. Cortex, 40(1), 19–40.
- Nisbett, R. E., & Wilson, T. D. (1977). Telling more than we can know: Verbal reports on mental processes. Psychological Review, 84, 231–259.
- Overgaard, M (2004): On the naturalising of phenomenology, Phenomenology and the Cognitive Sciences 2004;3(4):365-379.
- Overgaard, M. (2006): Introspection in science, Consciousness and Cognition, 15, 629-633
- Overgaard M, Gallagher S, Ramsøy TZ (2008): Integration of First-Person Methodologies in Cognitive Science, Journal of Consciousness Studies,15:5, 100–120
- Persaud, N., McLeod, P. & Cowey, A. (2007): Post-decision wagering objectively measures awareness, Nature Neuroscience, 10, 257-261
- Petitmengin, C., Navarro, V. & Le Van Quyen, M. (2007): Anticipating seizure: Pre-reflective experience at the center of neuron-phenomenology, Consciousness and Cognition, 16 (3), 746-764
- Ramsøy TZ, Overgaard M. (2004): Introspection and subliminal perception, Phenomenology and the Cognitive Sciences,3(1):1-23.
- Schooler, JW (2002): Re-representing consciousness: Dissociations between consciousness and meta-consciousness. Trends in Cognitive Science, 6, 339-344.
- Varela, F. J. (1996). Neurophenomenology: A methodological remedy for the hard problem. Journal of Consciousness Studies, 3(4), 330–349.
- Varela, F.J. & Shear, J. (1999): First-person methodologies: What, Why, How?, Journal of Consciousness Studies, 6 (2-3), 1-14
- Wundt, W. (1907): Über Ausfrageexperiments und über die Methoden zur Psychologie des Denkens, Psychologische Studien, 3, 301-360
- Anil K Seth (2007) Models of consciousness. Scholarpedia, 2(1):1328.
- Author (Year) Title Publisher