Memory Models

The Memory System: Working Memory

The memory system stores and organises sporting information gathered during information processing, using several interlinked stores. The working memory model (Baddeley and Hitch) describes short-term storage and active processing of information, made up of a central executive plus three sub-systems:

• Central executive – the control centre of the working memory; oversees and coordinates the flow of information between the other sub-systems and directs attention
• Phonological loop – deals with auditory (sound-based) information, e.g. verbal instructions from a coach or a teammate's call
• Visuospatial sketchpad – temporarily stores visual and spatial information, e.g. the flight path of a ball or the positions of other players
• Episodic buffer – links and integrates sight, sound and movement information into ordered sequences, and passes these on to long-term memory as the starting point for movement

The working memory has a limited capacity and can only hold information for a short period (around 30 seconds) unless it is rehearsed or acted upon. Information judged relevant is compared against memory traces held in long-term memory; irrelevant information is quickly discarded.

Long-Term Memory and Storing Information

Long-term memory (LTM) receives information from the working memory and has an almost unlimited capacity to store motor programmes/skills, often over very long periods of time. A well-learned skill is stored in the LTM as a motor programme – a set of stored instructions specifying the components needed to perform that skill.

Several strategies can help a performer store and later recall information more effectively:

• Association – linking new information to skills, actions or emotions already stored in memory, making it easier to recall
• Chunking – breaking a longer, more complex sequence of actions down into smaller, manageable sub-routines, making the whole sequence easier to learn, store and recall
• Rewards – reinforcing correct actions with praise or reward increases the likelihood that information about that action is retained
• Focus/concentration – reducing distractions and focusing attention on the task improves how well information is encoded into long-term memory
• Repetition – repeatedly practising an action strengthens the motor programme stored in the LTM, making it more automatic and reliable to recall
• Mental practice – rehearsing a skill in the mind, without physically performing it, can also reinforce the stored motor programme

Schema Theory (Schmidt)

Schema theory (Schmidt) suggests that, rather than a separate motor programme being stored for every possible variation of a skill, performers store a generalised set of rules, called a schema, that can be adapted to fit new but related situations. A schema is built from an existing motor programme, updated using feedback from past experience of performing the skill.

Each schema has four parts, split into two sub-schemas used before and after the action:

• Recall schema (used before/during the action, to initiate movement): 1) Initial conditions – information from the environment about the situation before the action begins (e.g. the position of the performer and other players); 2) Response specifications – information about what needs to be done (e.g. how much force or what type of action is required)
• Recognition schema (used during/after the action, to control and evaluate movement): 3) Sensory consequences – information from the senses about what the movement felt, looked or sounded like as it was performed; 4) Response outcome – feedback on how successful the action was in achieving its goal

Because a schema is a generalised set of rules rather than one fixed programme, it allows a performer to adapt a learned skill to new situations – for example, adjusting the run-up, grip and arm action used in a javelin throw to instead throw further or shorter, or with a different implement.