Cognitive style and learning strategies

1. Theoretical background: nature or nurture?
2. Pedagogical implications
3. Cultural influences and imitations
4. Conclusion
5. References

1. Theoretical background: nature or nurture?

 Before the 1970s, individual differences had been synonymous with differences in ability (Willing 1988:35), at least in the field of learning theory. Nevertheless, many psychologists in the 1950s and 1960s became increasingly concerned about the narrowness of abilities measured by standard intelligence (IQ) tests. Emphasis on abstract logical reasoning seemed to restrict intelligence to "convergent thinking" towards pre-determined answers but excluded the type of "divergent thinking" which leads to imaginative or creative innovation. Guildford (1965) introduced a model of the structure of the intellect in which he differentiated between a number of cognitive operations including convergent and divergent thinking (Lovell 1980:104). Divergent thought soon became equated with creativity, but although his (1975) concepts of fluency, flexibility and originality are still widely used, the value of his contributions to the understanding of creative thinking is now thought to be questionable (Ochse 1990:205).

 The real value of Guildford’s distinction was realised by Hudson (1968) who suggested that tests of divergent thought were not so much a measure of creativity as a sampling of the individual’s preferred style of thinking (Lovell 1980:105). From a study of sixth form science and arts students, Hudson found that science students, specially those specialising in physics, tended to prefer a convergent style of thinking and saw themselves as basically cold, dull and unimaginative. Similarly, arts students, particularly those specialising in English literature, history and modern languages, were more likely to be divergent thinkers and saw themselves as warm, imaginative and exciting but at the same time lacking in manliness and dependability (Lovell 1980:105).

 Hudson’s work was important in that it also showed a connection between style of thinking (or cognitive style) and the learners’ social behaviour and self-image.

 Hudson (1968) also found a relationship between convergent/divergent thinking and another bi-polar dimension known as syllabus-bound and syllabus-free orientation. Convergent thinks or "sylbs" were typically concerned with getting good examination marks and happily accepted the restrictions of a formal syllabus. "Sylfs", on the other hand, had intellectual interests that extended far beyond the syllabus, which they often found constricting (Lovell 1980:105). Parlett (1969) found that "sylbs were exam-oriented but had little personal interest in the subjects they studied. Although they were "model" students at university, attending more lectures, working harder and achieving higher marks in exams, "sylbs" were less successful than "sylfs" when it came to independent project work (Lovell 1980:106).

 Again, the distinction between "sylbs" and "sylfs" was not just limited to cognitive behaviour but included social and affective characteristics. Another study of sixth-form students, this time by Josephs and Smithers (1975), showed that "sylbs" tended to be "more conservative, controlled, conscientious and persistent, shy, cautious and practical "when contrasted with "sylfs". They were more intolerant and authoritarian in their outlook and more dependent upon their social group (conformists) (Lovell 1980:106).

 As many as 19 different ways of describing cognitive style have been identified, all of which consist of bi-polar distinctions similar to those described above (Entwistle 1988:47). All of these tend to be assimilated to the construct field-dependence-field-independence (Willing 1988:41), which has become a sort of general theory of perception, intellect and personality. Berry (1981) characterises this dimension as follows:

"The central feature of this style is the "extent of autonomous functioning" (Witkin, Gooddenough and Otman 1979); that is, whether an individual characteristically relies on the external environment as a given, in contrast to working on it, is the key dimension along which individuals may be placed. As the name suggests, those who tend to accept or rely upon the external environment are relatively more Field Dependent (FD), while those who tend to work on it are relatively more Field Independent (FI)" (quoted in Willing 1988: 41-42).

 Berry goes on to explain that individuals have a characteristic "place" on this dimension but that this may change according to circumstance and in response to specific training (ibid: 42).

 A summary of the findings of cognitive style research as they relate to the two contrasting poles of the field independent (analytical/field independent (concrete) dimension is given below in Fig.1.

Fig 1 Contrasts on the two poles of the Field Independent (Analytical) Field Dependent (Concrete) Dimension (from Willing, 1988)

Information processing

Learning strengths

 Human relations

  In order to understand better the notion of field dependence/field independence, it is worth explaining how the original distinction came about and how it differs from an alternative but complementary explanation of the source of cognitive style differences, namely the split nature of the brain.

 Witkin et al (1954) found that people differ from each other in the way they perceive both their environment and themselves in relation to it. Their original findings were based on the contrasting ways in which individuals establish the upright in tests involving tilted frames or tilted rooms. Field-dependent people tended to rely upon visual information from the outside world (hence the term field-dependent) whereas field independent people relied almost exclusively on internal cues such as muscle tension or sensations from the vestibular system in the ear (Lovell 1980:107) and ignored external evidence to the contrary. A brief description of these experiments is given in Witkin (1969:288-291).

 Later, an alternative (and simpler) way of measuring field dependence – field independence (FD-FI) was developed which consisted of having people pick out simple figures from a more complex design. Again, individuals were asked to deal perceptually with items in a field. For some (FI) people the simple figure almost "popped out" of the complex design, while other (FD) people were unable to find it even in the five minutes allowed (Witkin 1969:292).

 Witkin (1969:294) argues that "the style of functioning we first picked up in perception (…) manifest itself as well in intellectual activity". Field dependence or field independence are the perceptual components of a particular cognitive style. Thus "at one extreme there is a tendency for experience to be diffuse and global; the organisation of a field as a whole dictates the way in which its parts are experienced. At the other extreme the tendency is for experience to be delineated and structured; parts of a field are experienced as discrete and the field as a whole is structured" (ibid: 294).

 While scores for any large group of people on tests of FD-FI show a continuous distribution (ibid: 294). Witkin repeatedly found sex differences with females tending to be more FD and males correspondingly more FI. (Later studies, however, show the evidence to be conflicting – see Willing 1988:103.) Witkin attributed this discrepancy to different styles of child rearing. Thus he claims, for example, that mothers of field-dependent children tend to represent the world to their children as uniformly dangerous and satisfy all their children’s needs in the same way (e.g. a mother might breastfeed her baby every time it cried). Mothers of field-independent children, on the other hand, are more likely to specify sources of danger selectively and to respond differently needs. According to Witkin, the extent to which the mother articulates such early experiences determines the child’s later position on the FD/FI continuum (Witkin 1969:312).

 But just as there is a nature-nurture debate with regard to the source of intelligence differences, so differences in cognitive style can also be attributed to genetic factors. An alternative explanation is that cognitive style reflects the individual’s preferential use of one or other hemisphere of the brain much in the way that left-or right-handedness does. Evidence from brain research suggests that one gene determines the dominant hemisphere of the developing brain, while another relates to "handedness" (Entwistle 1988:48). While the specialisation of functions is relative rather than absolute (ibid: 48) and, in normal functioning, the two halves cooperate very closely to produce a unity, Levy (1979) argues that a perfect balance of strength only exists in about fifteen per cent of normal people: in all other cases, hemisphere strengths are unbalanced (Willing 1988:45).

 There is no room here to go into the question of hemispheric specialisation in any great depth, but Hartnett (1981) states that:

 "Recent brain research … provides evidence that the left cerebral hemisphere is specialised for logical, analytical, linear information processing, and the right hemisphere is specialised for synthetic, holistic, imagistic information processing. This evidence seems to parallel research on dual cognitive style models such as field independent/field dependent …, analytical/rational …, serialist/holist … and sequential-successive/parallel-simultaneous". (Quoted in Willing 1988:46).

  2. Pedagogical implications

 What are the implications, then, of cognitive style for the development and use of learning strategies? As mentioned above, the construct FD-FI has over the years become very broad and encompasses not only cognitive and metacognitive elements but also the socio-affective side of the learner. In order to avoid too much repetition, the socio-affective implications of learning style will be discussed in a later article that deals with personality. Here we shall refer to a more limited version of the FD-FI dichotomy which was developed with special reference to education and which according to Lovell (1980:106) has special significance for an individual’s choice of learning strategies although Lovell himself gives no examples. This is Pask’s (1969) distinction between serialist and holist styles of learning.

 A holist style involves a preference for setting the task in the broadest possible perspective and gaining an overview of the area of study so that the details are contextualised (Entwistle 1988:61-62). This has implications for metacognitive strategies such as previewing, organisational planning and directed and selective attention. Previewing will tend to come naturally but may be rather indiscriminate. It is perhaps more difficult for holistic to extract the organising principle from a text without explicit cues. Holists may have more difficulty in attending to task or deciding what is essential in the early stages. On writing task, they are more likely to discover what they want to say through a global strategy of drafting and redrafting rather than filling in an initial outline, and their approach tends to be "idiosyncratic and personalised" (Entwistle 1988:62). They may have difficulty with evaluating form.

 Holists use visual imagery and personal experience to build up understanding. Drawing mind-maps using imagery and colour will be useful memory strategies for holists (see Buzan 1989:95). Creative elaboration (e.g. making up stories) and personal elaboration are also likely to appeal to holists. However, they may need to develop strategies that compensate for a natural tendency to over generalise and ignore important differences between ideas. Such attention-directing strategies are described by De Bono (1976) and include "thinking tools" such as listing other people’s points of view, arguments for and against a proposal etc.

 In contrast, a serialist style is described by Pask (1969) as step-by-step learning. The focus is narrow, with the student concentrating on each step of the argument in order and in isolation (Entwistle 1988:63). Serialists approach the study of new material by stringing a sequence of cognitive structures together and thus tend to be very intolerant of redundant information because of the extra burden it places on memory (Lovell 1980:106). They are likely to use planning and selective attention strategies too early in an attempt to limit the amount of information they have to deal with. On writing tasks, they may need to make a considerable effort to "brainstorm" for new ways of approaching a subject and are likely to have difficulty in evaluating content, which "tends to be carefully structured and clearly presented, but may be dull and humourless" (Entwistle 1988:63) and "lacking in personal interpretation or independent conclusions" (ibid)-

 Unlike holists, serialists are good at noticing even trivial differences but are poor at noticing similarities. Thus they may need to use elaboration strategies that emphasise relating different parts of new information to each other as well as relating information to personal experience. A caveat must be added here. As with the FI/FD dimension of which the serialist-holist forms a part, few people are totally serialist or holist in their approach. Pask found some students who were versatile: they were equally comfortable with either style and could use both as appropriate. Other students, however, showed a marked over-reliance on one or other of these styles which gave rise to characteristic pathologies of learning (Entwistle 1988:62). It is these individuals who are likely to prove the most impervious to strategy training.

3. Cultural influences and imitations

 Finally, there is the question of how cognitive style relates to cultural background. Witkin himself identified field independence with a higher and more advanced degree of autonomy and individualisation (Willing 1988:48). Subsequent research (Witkin 1977; Berry 1979,1981) has shown that in "loose" migratory, hunter-gatherer societies in which the individual typically works alone and depends upon a high degree of perceptual discrimination and autonomous decision-making, field-independence is favoured. But in more stable, sedentary or stratified societies (usually agrarian) with "tight" family and social networks, relative field dependence seems to be the norm (Willing 1988:48-49).

 Modern industrial societies, however, are more complex. On the one hand, they present many of characteristics of agrarian societies although the extended family is rare in Northern Europe and America. Yet it might also be predicated that education would tend to produce a more "analytical" mode of thinking (Willing 1988:102). In fact a study carried out in Australia by the Adult Migrant Education Service (AMES) has shown that at least as far as language learning is concerned:

 "(…) learning modes cut across age levels, both sexes, and all levels of previous education. To a considerable degree, learning preferences actually cut across all biographical variables – including ethnic group". (Willing 1988:151)

 Over eighty per cent of the participants in this study were from large towns (50,000+) or cities and belonged to a wide number of ethnic groups, both European and Asian (ibid: passim).

 Unfortunately, research has also shown that perceptually-based testing devices such as the Embedded Figures Test are not reliable when the tested group itself is multicultural (Willing 1988:44). Willing (ibid: 44-45) cites the example of obviously highly "analytical" students from certain Asian cultures that were slower and less accurate in responding than some Europeans who were in all other respects far less analytical and claims that it would be necessary to reposition the entire scale in order to permit comparison between cultures. The reason for this seems to be the cultural bias involved in tests containing abstract geometric patterns. (Highly educated Asians with long exposure to Western culture, however, are presumably less likely to misunderstand what is required of them). Curiously enough, the Embedded Figures Test correlates quite highly with another culturally biased instrument – the standard IQ test in the low and medium range of the scales although not at the higher end (see also Skehan 1989:114-115 on FI as a disguised measure of intelligence).

In conclusion, cognitive style, in particular the FI-FD dimension, is a well-researched construct that includes not only cognitive and metacognitive elements but also the socio-affective side of the learner. Unlike Gardner’s (1984) theory of multiple intelligences (MI), it does not assume that linguistic functioning is separate from other types of functioning, but rather that people fall on a continuum between serialist and holist, analytical and intuitive, and independent and social, and have different strengths and weaknesses. Similarly, unlike MI theory, it does not presuppose that a person’s linguistic ability is more or less pre-determined at birth (see my article in Share No. 68) but rather that cognitive style is the result of complex interactions between hemisphere strengths and early learning experiences. Also, the fact that the Embedded Figures Test has been shown to be culturally biased should make us wary of assuming that hemisphere strengths are totally ‘biological" or "genetic" in origin. Above all, the research on cognitive style provides us with a rationale for diagnosing individual weaknesses, while suggesting that the ideal balance is somewhere in the middle of the FI/FD continuum. In this way, learners can be taught compensatory strategies so as to get the best of both worlds.

 © Douglas Andrew Town 1993, 2003

 References

Buzan, T. (1989). Use your head. London BBC Books.

De Bono, E. (1976). Teaching thinking. Pelican Books.

Entwistle, N. (1988). Understanding classroom learning. London: Hodder and Stoughton.

Gardner, H. (1984). Frames of mind. London: Heinemann.

Lovell, R.B. (1980). Adult learning. London: Croom Helm.

Oche, R. (1990). Before the gates of excellence: The determinance of creative genius. Cambridge: CUP.

Skehan, P. (1989). Individual differences in second language learning. London: Arnold.

Willing, K. (1988). Learning Styles in adult migrant education. NCRC Research: Adelaide.

Witkin, H.A. (1975). ‘Some implications of cognitive style for problems of education’. In Personality and learning 1. Ed. by Whitehead, J.M. London: Hodder and Stoughton.

 Douglas Andrew Town BSc (Hons) Psychology, MA (English Language Teaching), Diploma in  Translation (Spanish)  Profesor de la Universidad de Belgrano, Argentina (Licenciatura en Inglés).