Stimulus-response relationships, which have characterized the behaviourist perspective, could only explain what could be seen externally but not the measurable aspects that occur in the brain systems such as information processing, memory and perception. Earlier research and experiments in this field have found that animals could form generalizations out of the learned responses to stimuli other than those to which they had already formed a stimulus-response relationship (Littleton, Toates and Braisby, 2002). For example, Mercado, et al (2000) as cited in Littleton, Toates and Braisby, (2002) discovered that bottle-nosed dolphins who have been conditioned to distinguish pairs of shapes as either the same or different, could also apply this learned response to other objects with unfamiliar shapes. The most significant of these studies concerning the cognitive perspective, however, had been those dealing with the aspect of category learning. This type of learning suggests that people form their own schema or categories of things characterized by perceived relationships so that whatever information is received can be organized under these categories. For example, if we taste a certain type of exotic food and find, through our senses that it does not appeal to our taste or to our smell, we place that particular type of food under our category of 'bad-tasting food'. Thus, categories are likened to "formed" concepts about the world which enables us to judge, plan, anticipate or react to experiences. In turn, categories are formed through mechanisms that enable people to acquire them. One such mechanism, as demonstrated by the studies of Bruner et al (1956) as cited in Littleton, Toates and Braisby, 2002, is the hypothesis formulation and testing. Hypothesis is essentially a proposed explanation for an observed or perceived behaviour or experience that may be proven correct or otherwise. In showing how this works, Bruner et al conducted experiments that were called "successive scanning" where participants were asked to formulate one hypothesis at a time to solve mental puzzles and prove or disprove it through trial-and-error. This strategy was referred to as 'selective scanning'. Murphy and Allopena (1994) as cited in Littleton, Toates and Braisby, (2002) modified Bruner's experiment by assigning meaningful attributes to the objects for categorization in relation to a theme. They were able to show that background or prior knowledge plays a significant role in category learning.
In contrast to the hypothesis-testing and category learning views, two other cognitive psychologists, Fodor and Chomsky (1980) cited in Littleton, Toates and Braisby, (2002), argued that knowledge about categories, is actually innate. As such, the inductive problem was raised which rejected the notion of categories as being learned. Littleton, Toates and Braisby, (2002) suggested that a response to Fodor and Chomsky's argument could be that there are different aspects in learning and that categories could be formed through other means aside from hypothesis-testing.
Veering away from the mental and covert processes that attempted to explain the concept of learning, the Socio-cultural