Argument Theories of Learning - Toulmin's Argument Schema

Argument theory has been significant in studies of Computer Supported Collaborative Learning (e.g Andriessen et al, 2003: Ravenscroft & McAllister, 2006). The roots of argument theory lie in philosophy and conceptual analysis rather than in empirical educational research and it focuses on the structure and syntax of arguments. As I read and researched online learning tools I became aware of the need to also include the concept of argumentation and its related theories as these generally underpinned the tools specifically developed for an online discussion (de Laat & Wegerif 2007). The tools are generally created from the point of developing educational discussion whose affordance is supposed to be in encouraging critical and creative thinking. This also links to the theory of learning as discussed in the earlier blog on social learning theories as the discourse is central to the process here.

Toulmin’s Argument Schema

Toulmin’s explanation of informal argument has been very important in education, as it offers a good illustration of this approach to argument which treats it as if it was a kind of ‘grammar’ (Toulmin, 2003, updated edition). Criticizing over-formal and over-abstract accounts of good and bad arguments in terms of logical syllogisms rather than ordinary language, Toulmin introduced a description of what he called ‘informal logic’. His terminology such as ‘warrant’ suggests that his perspective was influenced by the kind of argumentation used by lawyers in courtrooms. According to him, argumentation can be broken down into the following moves:

• A claim states the standpoint or conclusion, for example: “The Kyoto protocol to reduce global warming is necessary.”

• The data are the facts or opinions that the claim is based on, for example: “Over the last century, the earth’s temperature has been rising as a result of greenhouse gas emissions.”

• The warrant provides the validation for using the data as support for the claim, for example: “Scientists agree that there is no other explanation for this rise in temperature.”

• Optionally, the backing provides specific information supporting the warrant, for example: “Scientists have identified the atmospheric mechanisms whereby greenhouse gases cause a warming of the earth’s surface.”

• A qualifier adds a degree of conviction to the conclusion, indicating the degree of force, which the arguers attribute to a claim, for example: “However, the earth’s temperature has been found to fluctuate over geological time, in some cases without any obvious cause.”

• Exceptions to the claim are expressed by a rebuttal, for example: “The Kyoto protocol would not be necessary if the world’s countries voluntarily reduced their output of greenhouse gases.” (Examples taken from Andriessen, 2006)

This theory can be seen underlying the tool called Araucaria (to be explained in a later blog) where the users can develop Toulmin Diagram to display their understanding of a topic. Therefore, this theory can be seen to be basic to the communication within a group when discussing specific issues. The elements of the argument from each participants view point remain the same when using a linear text display tool or a graphically and spatially displayed text tool. This approach allows the thinking skills to be developed as these are central to learning online.

Social Learning Theories: Socio-Cultural Theory

Individuals also can come together for the need to learn and form community of learners. Community of learners, according to Brown and Campione (1994) is a pedagogical model that intends to take advantage of the distributed expertise and cognitive diversity. ‘The approach is focused on adopting the goals, values, beliefs, and forms of discourse characteristic to scientific practice.’ (De Laat 2006, pp 22) Conceptual advancement is made by cultivating each member’s own expertise with the participants engaging in a self-regulated and collaborative inquiry as a group (Lehtinen et al., 1999 cited by De Laat 2006). The participants are apprentice learners, learning how to think and reason in a variety of domains (Brown, 1997). In a community of learners they try to foster support overlapping zones of proximal development that stimulates growth through mutual appropriation and negotiated meaning.

Hence socio-cultural learning theory takes into consideration that learners will achieve more with assistance from others, which according to Vygotsky (1978, and also cited by Wells 2000) is the ‘zone of proximal development’. Lave and Wenger (1991) find that there are three characteristics in relation to the interpretation of the concept of zone of proximal development. They are :

• There is a distance between the abilities of problem solving of a learner who works alone and the abilities displayed by a learner who works in collaboration or is assisted by other more experienced / knowledgeable people. The idea being that learning is increased through support and from scaffolding between learners and experts.
• There is a distance between the cultural knowledge that comes from historical contexts or events and the day-to-day experience of individuals. Here Lave and Wenger (1991) cite Hedegaard (1988) who explains that there is distance between understood knowledge, which is transmitted via instructions and active knowledge, which comes from participation and thus owned by the individual. 
• Final interpretation is from Engestrom (1987, cited by Lave and Wenger 1991) who states that there is a distance between the everyday actions of individuals and new forms of societal activity that will transform the society. The focus is on the change / transformation that will occur in the culture as individual becomes functioning members of the community.

Learning is about actively engaging in the practices of the community. This means learners will eventually become full participants in the community of practice, this in itself refines the practices of that community and the interconnectedness of communities allows the whole organisation to become more effective and valuable within the society. So learning is situated in the activity an individual carries out, when trying to become a member of a community. This according to Lave and Wenger (1991) is the central characteristic of legitimate peripheral participation, ‘a way to speak about the relations between newcomers and old-timers, and about activities, identities, artefacts, and communities of knowledge and practice’ (pp. 83).

Newcomers work with experts or old-timers in specific activities. The newcomer assists in and later carries out the activities peripheral to the central activity. The old-timers acts as a role model as well as provides explanation and guidance to the newcomer who takes on increasingly more responsibilities and finally to carry out the entire activity alone. This brings back the old idea of apprenticeship model of learning, though there is definitely more to this form of learning than just watching and copying as it ‘emphasises on comprehensive understanding involving the whole person rather than receiving a body of factual knowledge about the world; on activity in and with the world; and on the view that agent, activity, and the world mutually constitute each other’. (Lave and Wenger 1991, pp. 85)

After reading about the socio-cultural theory I can see elements that can be linked into the way a community of practice learns. These elements are related to learning from each other and learning to think with each other within synchronous online learning discussion. The community of learners are therefore just another kind of community and carry the familiar characteristics of common need, focus on sharing resources and experiences to support and scaffold learning leading to development of self and the community. Therefore the learning is individual as well as in the group. Thus the old apprenticeship model within professions have a basis within well explained learning theory.

Social Learning Theory - Communities of Practice

Wenger (1999) explains that since humans are social beings learning occurs in the social context. The experiences and activities we engage in are seen to be meaningful and allow us to participate in the world. This active participation takes place through knowing and the ‘knowledge is a matter of competence with respect to valued enterprises’. (Wenger 1999, pp. 4) Wenger (1998) explains learning is based on the following premises –

1. ‘We are social beings. Far from being trivially true, this fact is a central aspect of learning.
2. Knowledge is a matter of competence with respect to valued enterprises – such as singing in tune, discovering scientific facts, fixing machines, writing poetry, being convivial, growing up as a boy or a girl, and so forth.
3. Knowing is a matter of participating in the pursuit of such enterprises, that is, of active engagement in the world.
4. Meaning – our ability to experience the world and our engagement with it as meaningful – is ultimately what learning is to produce.’ (pp 4)

Learning is an integral part of our life and it takes place through participation in communities. When we get placed in situations that challenge our ability to respond we seek out others who can enable us to learn and develop new practices. This allows for development of individuals and leads to their becoming fully functional members of the society in which they live. This makes it necessary to look at individuals and also the social and material environment with which they interact, as a part of their development. Learning takes place through people’s engagement in actions and interactions which is embedded in culture and history. These actions and interactions allow learning to take place and this in turn replicates and transforms the social make-up in which it occurs. (Wenger 1998)

Learning shapes the individuals through the activities of the past generations. The process also leads to change / progress in the society. To learn, the individual first has to make sense of others and then construct knowledge, which will allow him to relate to self and to his situation. The emphasis is on the interaction with others, which leads to becoming functioning members of community before internalisation of knowledge can occur. This indicates that thinking of individuals is influenced by the group. Thus people with common interest or activity will come together to form a community of practice. Within this group mediated understanding there is a merger of social and intellectual processes and a tension between the collective understanding of the group and the individuals within. 

De Laat (2006) identifies this tension as the driving force for the collective processing of the group. The example he gives is, ‘when an individual member of the group expresses his or her opinion in relation to the shared public understanding of the group, this will be based on an attempt to synthesise his or her own understanding with the public one. The other members of the group will compare this new synthesis with their own understandings of the group-accepted version and their own disagreements with it. Depending on the outcome of this process there may be further interaction and negotiation until the group accepts a new meaning or understanding.’ (pp. 16)

A community of practice according to Wenger (1998) has three dimensions which would be relevant when learning in groups is considered:

• It is a joint endeavour as understood by the members. It can be also continually renegotiated by its members.
• The members form a social entity through relationships of shared engagement.
• The members share resources that they have developed over time so that this becomes a repository which can be accessed by any group member.

The issues central to this social learning theory are related to learning in groups, having a discussion where participants describe their experiences and understanding, comment of others contributions in an effort to scaffold, collaborate and negotiate meaning. I believe this helps them to reflect and internalise meaning and build on their understanding of the concept being discussed. These have to be central in any elearning development as edcucationalists continuously endeavour to ensure that learners gain from learning with and from each other.

Understanding Behaviourist Learning Theories

Recently Merriam and Caffarella (1991, cited by Jarvis 2000) explain that learning theories can be classified into four types –

•             behaviourist,

•             cognitive,

•             humanist and

•             social/cultural. 

However, it must be kept in mind that not all theories can be placed discretely into a category and that overlap is seen clear when discussing the theories later within this section. Also no single theory can provide us with an explanation of how individuals learn. Therefore understanding of the various theories will allow us ‘an opportunity to examine or re-examine our own beliefs and assumptions about people, knowledge, motivation, environment, assessment and many other factors associated with learning’ (Dean & Kenworthy 2000, pp. 46)

Behaviourism

Learning is very much investigated by psychologists in educational and experimental psychology. Whether we regard life in terms of a community or an individual, it relates to learning. Many of the psychological learning theories indicate that individuals are shaped or moulded by a variety of sources external to him. These theories have been derived from experimentation on animals and not humans therefore they are limiting. Some of the behaviour was also influenced by use of reward to encourage teacher’s desired objectives. These formed the basis of teaching children as well as adults thus there has been a tendency to over emphasise the behavioural approach. These theories are also unsatisfactory, as they do not consider the ‘self’ within them. In the 1800’s, the clinical psychologists did experiments on animals. They were based on either ‘trial and error’ or ‘stimulus – response’ principles.

The school of behaviourism believes that learning is a change in observable behaviour, which arises from a connection between stimulus and response. ‘By manipulation of this link the behaviour can be altered’ (McKenna 1995). Gagne’s (1985) definition of learning represents this approach clearly, ‘a change in human disposition or capability that persists over a period of time and is not simply ascribable to process of growth.’ McKenna (1995) explained the early theory, proposed by Watson, that behaviour was learned when stimulus and response are linked frequently. This was also linked to Guthrie’s theory of contiguity where ‘if a stimulus occurred at the same time as response, the response would be repeated when the stimulus subsequently occurred again.’ (McKenna 1995 pp 29) Behaviour theorists find learning to be an acquisition of new behaviour and nothing else. This, however, does not take into account all kinds of learning, as activities of the mind are not addressed. It also does not explain how some learning, like recognition of new language by children occurs, where there is no reinforcement involved. Cust (1995) criticises that mental activity or thinking is not regarded as important by behaviourists. They believe that learning is a response to the environment around the individual, which leads to building of behaviour by contiguity, repetition and reinforcement.

Connectionism

This was one of the first theories to gain recognition. Thorndike (1928) used the trial and error method to form a learning theory. He imprisoned a hungry cat in a specially constructed cage / puzzle box. A dish of food was placed outside the cage just beyond the reach of the cat. The cat could escape from the cage only if it succeeded in operating a specially constructed latch. First the cat clawed at the bars of the box – random movements that contributed to its escape – by chance it succeeded in operating the latch and escape to enjoy the food. Over a number of trials Thorndike found there was a gradual decrease of errors. Till eventually success! The cat went straight to the latch and the food.

Thorndike stresses the importance of the three laws regarding the learner’s efforts being followed by success. They are :

Law of Effect :

The bond between the stimulus and response is weakened or strengthened because of the level of emotive satisfaction that accompanies the action. The greater the pleasure obtained from the learning experience, the more learning will occur. If the learning does not produce pleasurable experience it will not produce learning. Therefore the emphasis is on the importance of the end effect of the response (McKenna 1995).

Law of Exercise :

Repetition of meaningful actions results in substantial learning. The more meaningful the act to the learner, the greater will be the resultant learning. Thus practice makes perfect, provided the action is seen to be meaningful. This is very much in support of the suggestions of Watson as discussed by McKenna (1995).

Law of Readiness :

If the organism is ready for the connection then the result is pleasurable and learning is increased, if not, then the opposite result occurs. This relates to both physiological and psychological readiness. For example, a child cannot be toilet trained unless his nervous system is sufficiently developed. In higher education, the learner will find it hard to grasp the subject matter unless he has understanding of the foundations of the subject from his school education.

He proposes ‘that is the learner discovers some act or explanation to be effective or valid it will be repeated until such time as the consequences of the action no longer produce the desired or expected results’. (cited by Jarvis 1992, pp 76) Thus the theory started with the learner – something that is very useful when considering adult learning. In fact all his three laws have an implication on the learning theories related to adult learning.

Conditioning

In contrast to the above, theories of conditioning started with the teacher. Ivan Pavlov (1849 – 1936) a Russian psychologist developed an extensive theory of conditioned reflexes called ‘Classical Conditioning’. It is exclusively concerned with reflexes or automatic responses, which are triggered by specific stimuli.

Pavlov used the salivary reflex – this is triggered by the sight or smell of food and is an automatic response to such a stimulus. Response refers to any action on the part of the organism. Salivary reflex response is an Unconditional response – a natural response, as it is triggered by an Unconditional Stimulus – a natural stimulus.

Food (unconditional stimulus)  Salivation (Unconditional Response)

If a buzzer (Conditional Stimulus) is used in conjunction with the food, the response remains the same. Repeated over a period of time it was found that when the buzzer (Conditional Stimulus) was activated, it was associated with food (Unconditional Stimulus) and salivation (Conditional Response) was produced. Thus the learner learns / is conditioned to associate the presentation of the reward with the stimulus that occurs just prior to the reward being presented. The response once conditioned could be produced by other stimuli similar to the conditioned response. Pavlov called this concept generalisation. Both Woolfolk & Nicolich (1980) and Lovell (1987) are of the opinion that conditioned responses and the emotional component of our attitudes are formed through classical conditioning.

Skinner (1951, cited by Gross 1992) performed his experiments on rats and pigeons. His theory is called ‘Operant Conditioning’, where the response is shaped by the reward. Every action that approaches, approximates or achieves the desired behaviour of the learner is rewarded. He also concurs with Thorndike that the end result of behaviour was important in the repetition of that behaviour and not just random response to a stimulus. The repetition of behaviour can be increased or decreased through positive or negative reinforcement. This means learning can be manipulated.

It is based on – learning at all stages of life. Children must learn social habits acceptable to the community. Adults must learn how to perform their jobs and how to meet the responsibilities of family life. Daily life is about learning as life is a succession of major and minor problems that have to be solved by learning.

Skinner used reward and punishment to influence behaviour. He stated that for any behaviour that is a Conditioned Response, to continue, it was necessary to reinforce the Conditional Stimulus by giving a reward and vice versa, if an alteration in behaviour is required. When learning a new behaviour the reinforcement has to be frequent but to maintain the behaviour intermittent reinforcement is sufficient. It is also important to remember that not all conditioning maybe intentional, or conscious, but that it occurs definitely during the process of education of adults. (Jarvis 1992) This type of learning can be seen during the process of socialisation within a workplace. The reward is being accepted within the work team and this allows for the individual to continue learning in the workplace.

Social Learning Theory

The social learning theory of Bandura emphasizes the importance of observing and modelling the behaviours, attitudes, and emotional reactions of others. Bandura (1977, pp 22) states: "Learning would be exceedingly laborious, not to mention hazardous, if people had to rely solely on the effects of their own actions to inform them what to do. Fortunately, most human behaviour is learned observationally through modelling: from observing others one forms an idea of how new behaviours are performed, and on later occasions this coded information serves as a guide for action." Social learning theory explains human behaviour in terms of continuous reciprocal interaction between cognitive, behavioural, an environmental influences. 

The component processes underlying observational learning are: 

• Attention - the individual notices something in the environment (this relates to modelled events like distinctiveness, affective valence, complexity, prevalence, functional value, and observer characteristics like sensory capacities, arousal level, perceptual set, past reinforcement).
• Retention - the individual remembers what was noticed (this relates to symbolic coding, cognitive organization, symbolic rehearsal, motor rehearsal).
• Motor Reproduction - the individual produces an action that is a copy of what was noticed (this relates to physical capabilities, self-observation of reproduction, accuracy of feedback).
• Motivation - the environment delivers a consequence that changes the probability the behaviour will be emitted again (this relates to external, vicarious and self-reinforcement).

Processes integral to Learning from Observation

     Attention

     Retention

Motor Reproduction

     Motivation

Figure 1

So Bandura's work draws from both behavioural and cognitive views of learning. He believes that mind, behaviour and the environment all play an important role in the learning process where the benefits of this process are :

1. The highest level of observational learning is achieved by first organizing and rehearsing the modelled behaviour symbolically and then enacting it overtly. Coding modelled behaviour into words, labels or images results in better retention than simply observing
2. Individuals are more likely to adopt a modelled behaviour if it results in outcomes they value.
3. Individuals are more likely to adopt a modelled behaviour if the model is similar to the observer and has admired status and the behaviour has functional value.

Information Processing Model

As suggested earlier, Gagne finds learning to be a change in human disposition or capabilities, which will persist over a period of time. Individual behaviour must change in order for learning to take place. So according to Gagne the behaviour of the individual, before and after the learning incident must be compared. He presents a teacher centred approach to learning. Teachers carry out the task of promoting learning by designing of instruction, managing of instruction, and evaluating student learning. The teacher plans and controls the events / stimulus, external to the learner. Therefore Gagne’s work is highly regarded in learning of psychomotor skills. He does not however regard the learner as completely passive, (Joyce & Weil 1986) for the learner has to make the needful connections in order to learn.

 Figure 2

Coulter (1990) describes Gagne’s learning process as :

1.            A stimulus from the environment is detected by receptors.

2.            It is sent on to sensory register.

3.            From there it is transferred to the short-term memory.

4.            Later it is moved to the long-term memory.

5.            Encoding takes place.

6.            Retrieval of information from here allows for recall.

7.            Response generator will identify appropriate action

8.            This response is related to the effectors.

Explained thus, it would appear that issues like motivation, goals, expectation and intentions of the individual are of lesser importance, opines Bigge (1982). (see figure 2)

Gagne has proposed 8 types of learning, of which 7 are placed in a hierarchy and the 8th, he feels, can occur at any level. The understanding of these levels of learning leads to the opinion that the information processing model of Gagne cannot be viewed entirely as a behaviourist approach.

             Signal Learning can occur at any level. It is a form of classical conditioning and is one of the ways in which people develop attitudes and prejudices throughout life.

             Stimulus – Response is the same as operant conditioning.

             Motor and Verbal Chaining – are at the same level. Motor chaining refers to learning of skills, while verbal chaining is rote learning. Practice is needed to achieve correctness. Reinforcement is necessary to ensure that acceptable sequence is maintained. This is synonymous with the trial and error learning as well as operant conditioning.

             Multiple Discrimination – relates to intellectual skills with the ability to distinguish between similar types of phenomena, which will allow the learner to decide which similar type is correct for any specific situation. In contrast to discrimination is the ability to classify.

             Concept Learning – Concepts are abstract notions which link similar phenomena together. As the ability to think in abstract starts from adolescence (Piaget cited by Gross 1992) the education / learning in adults will differ from learning in the children. This aspect is important and coincides with Knowles’s concept of Andragogy.

             Rule Learning – where the individual gains the ability to respond to signals by a number of responses.

             Problem Solving – is the highest order of learning and the individual is able to draw upon previously learnt rules to solve particular problem. This is similar to the cognitivism approach of this level of learning as pointed out by Woolfolk & Nicolich (1980) Simply applying the prior knowledge to solve a problem is not problem solving, for it leads to new learning.

Gagne (1985) draws on Dewey’s explanation of the problem solving approach. First, there is a problem presented to the learner. Next the learner, ‘defines the problem, or distinguishes the essential features of the situation’ (Gagne 1985, pp 178). The leaner will form a hypothesis to solve this problem from this definition. Next, this hypothesis will be verified by application, to see if the problem can be solved. If it does not then a new hypothesis will be formulated and tried. Thus the successful hypothesis is learned and new rules maybe formed.