IMPLICIT LEARNING
I bet everyone, at least once in a lifetime, dreamt about learning without effort. Lightly, pleasantly, knowledge enters our brain. More or less it is how the implicit learning look like. It is unintentional learning that proceeds without consciousness and cognitive control. It is possible because our brain works incessantly. The cognitive system keeps processing the stimuli, even when we are not aware of that. In every single moment we are just bombarded with an unimaginable amount of information from which we consciously register only a small fraction.
It was Reber (1967), who for the first time used this term and conducted first research in the field. The unintentional learning, also called implicit learning, is defined as a cognitive process via which a subject acquires implicit knowledge about the complex structure of the environment (Wierzchon, 2009). The acquired knowledge has three features: is abstract, is acquired without conscious control and can implicitly influence on problem solving and decision making.
To fully understand the phenomenon of implicit learning we have to know how does the cognitive system work. Surprisingly, our system seems to be lazy. It uses the “cognitive miser” rule, which means that it tries to achieve maximum results with minimum effort. To enable our daily functioning our cognitive system perceives, encodes, stores and retrieves all incentives, internal, coming from our body, as well as external. Because quantity of stimuli exceeds the possibilities of our system, it processes only part of them. The limits of our system cause selection and reduction of information, that are done via attentional mechanisms.
Regarding the limits of cognitive system and information processing two theories are of special interest: theory of non-specific resources (Kahneman, 1973; Navon, 1984; Baddeley, 2001) and theory of specific resources (Allport, 1980; Wickens, 1984; Hirst and Kalmar, 1987). The former assumes that exists one pool of resources, that can be used on different processes when it is needed. If the process is highly controlled, it consumes numerous resources and only simple, not requiring engagement actions can be taken simultaneously. Simply, there are not enough resources to conduct all the actions at one time. The latter theory states that resources are divided between modalities. Therefore, we can easily control an action absorbing sight and audition, but we will have problems to coordinate two actions using the same sense.
Leaving the differences between the theories aside, it is essential to know that the resources are limited. Our activity- learning and remembering- causes cognitive costs, which means using the resources. That is why the system tries to reduce those costs. The amount of resources usage depends on the level of the processed knowledge and on the level of process’ automatization. The rule is simple: the higher the automatization, the lower the costs. We can differentiate various levels of automatization: controlled processes (not automatized at all), automatized processes, which became quick and efficient by procedure of the explicit knowledge (a good example here is driving or walking) and processes originally automatic. According to Bargh (1994) the automatic process occurs when at least one of the following conditions is fulfilled. The process has to be: unconscious, unintentional, effortless or uncontrolled. A majority of scientists agree, that implicit learning is one of the originally automatic processes, fulfilling all four criteria. The controlled processes consume more resources, but can be easily modified. In contrary the automatic processes, once triggered, go to an end without any possibility of modification. Their advantage is that their are much faster, than controlled processes.
Research in the field of implicit learning is conducted from the second half of the XXieth century. Most commonly they are analyzed in context of cognitive development and language acquisition (Ellis, 1993). The scientists’ interest concentrates on: knowledge representation, level of consciousness and overall process characteristic, meaning its form, not substance. Other important questions are posed about cognitive costs, features of cognitive system and automaticity level.
The conducted research can be divided into three paradigms. The first and the most popular one is the artificial grammar learning (AGL), proposed by Reber (1967). The procedure is as follows: for two groups of the participants a series of letter strings are presented. One of the groups has to memorize grammatical strings, the other one- random ones. In the second stage both groups watch strings, from which half is grammatical and half is not. The task is to classify, whether the presented strings are grammatical or not. Group that was presented the grammatical strings in the first stage had significantly more correct answers, although the participants could not explain the grammatical rules used in the experiment.
The later research is a modification of the one described above. It was conducted by Reber (1976; Reber and Lewis, 1977; Reber and Allen, 1978). In this case both groups had to memorize the appearing letter strings, but one of them was overtly informed about the grammatical rule implemented in the strings. Participants from this groups had longer memorizing time, they retrieved less strings and committed more errors. The bold conclusion was: explicit processing is worse than the implicit one. Other research shed more light on that question (Dulany et al., 1984; Mathews, 1989) and the conclusion was made that participants that learnt the material in a implicit way are at least equally good in performing later tasks as people conscious of the grammatical rules.
The research was also conducted in the paradigm of controlling complex systems (Broadbent, 1977; Berry and Broadbent, 1984; Broadbent et al., 1986; Berry, 1984 and 1991; Stanley et al., 1989). The main conclusion was that participants are able to control the systems long time before they can verbalize the knowledge about them.
The last group consists of sequence learning research. Nissen and Bullemer (1987) used tasks that measured serial reaction time and Lewicki (1988)- matrix scanning task. The results showed that the reaction time was decreasing when sequence was being repeated, but when the sequence was random it was increasing. The results did not depend on whether the participants claimed to notice the sequence or not.
To sum up, implicit learning is a process of learning an ability or material unintentionally. The acquired knowledge is predominantly unconscious, however the learned material has to be consciously perceived. The process is automatic and does not consume the resources. During implicit learning our level of procedural knowledge increases. The acquired information is rather long term and forgetting depends on time and repeating the stimulus. Tries to consciously control the process of implicit learning does not influence its efficiency (Wierzchon, 2009).
Literature:
Allport, D.A. (1980). Patterns and actions: Cognitive mechanisms are content specific. [W:] W.G. Claxton (red.), Cognitive psychology: New direction. Londyn: Reutledge.
Baddeley, A.D. (2001). Is working memory still working? Unpublished Paper for American Psychologist.
Bargh, J.A. (1994). The Four Horsemen of automaticity: Awareness, efficiency, intention, and control in social cognition. [W:] R.S. Wyer, Jr., i T.K. Srull (red.), Handbook of social cognition (2nd ed.), Hillsdale, NY: Erlbaum, 1-40.
Berry, D. C. i Broadbent, D.E. (1984). On the relationship between task performance and associated verbalisable knowledge. Quarterly Journal of Experimental Psychology, 36, 209-231.
Berry, D.C. (1984). Implicit and explicit processing in the control of complex systems. Niepulikowana rozprawa doktorska. University of Oxford.
Broadbent, D.E. (1977). Levels, hierarchies and the locus of control. Quarterly Journal of Experimental Psychology, 29, 181-201.
Broadbent, D.E., FizGerald, P., Broadbent, M.H.P. (1986). Implicit and explicit knowledge in the control of complex systems. British Journal of Psychology, 77, 33-50.
Dulany, D.E., Carlson, R., Dewey, G. (1984). A case of synhetical learning and judgment: How concrete and hoe abstract? Journal of Experimental Psychology: General, 113, 541-555.
Hirst, W. I Kalamar, D. (1987). Characterizing Attentional Resources. Journal of Experimental Psychology: General, 116(1), 68-81.
Kahneman, D. (1973). Attention and effort, Englewood Cliffs, New Jersey: Prentice-Hall, Inc.
Nissen, M.J. i Bullemer, P. (1987). Attentional requirements of learning: Evidence from performance measures. Cognitive Psychology, 19, 1-32.
Reber A.S. I Allen, R. (1978). Analogy and abstraction strategies in synthetic grammar learning: A functionalist interpretation. Cognition, 6, 189-221.
Stanley, W.B., Mathews, R., Buss, R. i Kotler-Cope, S. (1989). Insight without awareness: On the interaction of verbalization, instruction and practice on a simulated process control task. Quarterly Journal of Experimental Psychology, 41, 553-577.
Wickens, C.D. (1984). Processing resources in attention. [W:] R. Parasuraman i D.R. Davies (red.), Varietes of Attention, New York: Academic Press.
Wierzchoń, M. (2009). Koszty poznawcze uczenia mimowolnego. Kraków: Wydawnictwo Uniwersytetu Jagiellońskiego.
