Combinations of Simple Mechanisms Explain Diverse Strategies in the Freehand Writing of Memorized Sentences
| Autor: | Peter C.-H. Cheng, Erlijn van Genuchten |
|---|---|
| Rok vydání: | 2018 |
| Předmět: |
Adult
Male Handwriting Adolescent Computer science Cognitive Neuroscience Short-term memory Experimental and Cognitive Psychology Models Psychological computer.software_genre 050105 experimental psychology Task (project management) Style (sociolinguistics) Young Adult Artificial Intelligence Humans 0501 psychology and cognitive sciences Control (linguistics) 050107 human factors Language business.industry Working memory 05 social sciences Cognition Memory Short-Term Task analysis Parallelism (grammar) Female Artificial intelligence business computer Natural language processing |
| Zdroj: | Cognitive Science. 42:1070-1109 |
| ISSN: | 0364-0213 |
| DOI: | 10.1111/cogs.12606 |
| Popis: | Individual differences in the strategies that control sequential behaviour were investigated in an experiment in which participants memorised sentences and then wrote them by hand, in a non-cursive style. Thirty-two participants each wrote eight sentences, which had hierarchical structures with five levels. The dataset included over 31 thousand letters. Despite the deliberately constrained nature of the task and stimuli, 23 patterns of behaviour were identified from the durations of pauses that occurred before the inscription of letters at four chunk levels, spanning letters, word, phrases and sentences. A critical path task analytic model, Graphical Production of Memorised Sentences (GPoMS), shows that the control of writing relies on cues that continuously switch between motor actions and chunk retrievals in a just-in-time fashion at the level of letter information. GPoMS explains the individual differences in terms of variants of a motor production mechanism and variants of a chunk retrieval mechanism, which involve varying degrees of parallelism between cognitive actions and motor actions. A graphical technique for constructing GPoMS models was developed that enabled the estimation of on-going working memory demands during production. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Plný text