What are the cognitive processes involved in responding to an assessment task?

Activation, selection and response formation

When a student encounters an assessment task, such as a question, their cognitive response is rapid. By understanding these cognitive processes, we might write better assessment tasks and exercise more caution in inferring what their response tells us about what they might know.

We might break the cognitive process of responding to an assessment task down into three stages: activation, selection, and response formation. We will consider these in turn.

Activation using cues

Activation is prompting goal-directed behaviour in the student towards responding to the assessment task.

Activation depends on cues that are written into an assessment item. Cues direct the student's attention towards the target knowledge held in long-term memory. A cue is a signal that, if effective, will trigger the retrieval of the desired memory trace. For example, in Religious Studies we may ask ‘How did the suffering seen by Siddhartha Gautama contribute to his belief in enlightenment?’. The cue words in this question are ‘suffering’, ‘Siddhartha Gautama’ and ‘enlightenment’, each of which may help students identify the body of knowledge they must draw upon to answer the question.

For cues to perform their purpose there must be a cue-target match¹, in other words the cues must be associated in the mind of the student with the desired memory traces. An incorrect or partial answer by a student may not be because they do not know the answer but instead due to cue activation problems. For instance, the cue word may not have been used by the teacher sufficiently when the memory was encoded². Alternatively, the cue word may be associated with more than one construct or schema, or the cues may be too broad for the student to accurately identify what knowledge they are expected to draw upon in answering the question.

Consider this alternative to the above question: ‘What role does suffering play in Buddhist belief?’ If the teacher expects students to talk about enlightenment or Siddhartha Gautama in their answer, it may be less likely that they do so than were they asked the question with more precise cues. The student may not lack the knowledge; it is just that the question does not trigger the required response. However, perhaps it is the strength of this schema – the ability of the student to make these connections without prompt – that you wish to test. How precisely we select cues is important.

Selection

Activation evokes an automatic process where the retrieval of knowledge by the student is triggered almost instantly by the cue/s (e.g. 2 + 2 = 4), but often this activation feels effortful, and we might better describe it as a search for knowledge which uses the cues as prompts.

Searching for knowledge is imprecise as cues can coactivate related knowledge: knowledge that is not precisely relevant to the test item. Unless the student has fluently identified the response needed, there will be further cognitive processes needed before a response can be given. Therefore, the second stage is the selection of relevant and important knowledge for responding. This is a discriminatory process whereby we make choices about what knowledge will be useful to us in responding to the test item. This process tests whether a student's knowledge is well-structured and codified, enabling the retrieval of relevant information. It is more likely to be successful if the cues resemble the way the knowledge was originally learned.

One important concept to consider is memory interference, where the brain cannot distinguish between related or competing memories, resulting in a risk of incorrect knowledge being retrieved. The knowledge must be integrated (connected to other knowledge in the schema appropriately), distinguishable from other knowledge (such that the student will not confuse one concept with another), and reflective of the subject architecture (ordered and structured in the desired way). Students will be more successful in their search for the answer if their knowledge is structured well in long term memory: it is stable and has structural integrity. The choice of cues is again important in this respect. Including discriminatory cues³ – ones for which there is only one possible related memory trace – will reduce the risk of memory interference. Conversely, providing too many cues may increase the pool of possible responses⁴, thereby increase the chance of irrelevant answers. In general, providing a limited number of cues, including at least one that is discriminatory, is a good bet.

Response formation

The third and final stage involves response formation by the student using the knowledge they have retrieved. This complex process includes ordering knowledge, shaping the response into a desirable form, and inhibiting task-inappropriate responses that may have been brought to mind in the search process. A student's ability to do this well depends on the state of their long-term memory and their speed of processing and working memory.

Students may provide an inadequate response in an assessment not because they do not possess the required knowledge but because they struggle to construct a response in the form expected. Teachers often use the term ‘exam technique’ to describe this skill. However, it may be more helpful to think of this ability as domain-specific: an ability related to the body of knowledge in scope rather than a general skill or set of techniques that students can master. Students will need practise in applying their knowledge with dexterity if they are to show us what they know in the form required.

1. Thompson & Tulving (1970), Associative Encoding and Retrieval: Weak and Strong Cues.

2. Tulving, E. (1983). Elements of episodic memory.

3. Eysenk (1979), Depth, Elaboration, and Distinctiveness.

4. Nairne (2002), The Myth of The Encoding-Retrieval Match.