Structured word-lists as a model of basic schemata: deviations from content and order in a repeated event paradigm

ABSTRACT

Repeated events are common in everyday life, but relatively neglected as a topic within memory psychology. In two samples of adults, we investigated memory for repeated, schema-establishing simple events (operationalised as structured word-lists), and the effects of deviations within those events. We focused on the effects of deviations from two core dimensions of schema: content and order. Across three successive word-list events, we established and reinforced a basic list schema by always presenting three content categories in the same order. These expectations were violated in a fourth and final word-list. We measured the effects on memory of both the violating and the schema-establishing lists in multiple recall attempts over a period of one month. We measured correct recall, misattribution errors, metacognitive awareness of list-organisation and deviations, and recall organisation. Across all delays and across all word-lists (not only the final one), content changes increased recall, whereas order changes decreased recall. Participants were also more aware of content changes than order changes. These disparate effects suggest that the two types of schema-deviations may have qualitatively different effects on memory for specific instances of a repeated generic event. Cognitive processes underlying memory for typical and exceptional instances of repeated events are discussed.

KEYWORDS:

• Schema
• schema-deviation
• repeated events
• recall organisation
• metacognition
• source memory

Acknowledgements

The authors would like to thank Marek Vranka, Pavel Humpolíček, Stanislav Ježek, Otto Eibl, and Roman Chytilek for providing laboratory space and equipment; Hana Řádová, Veronika Horníčková, Dominika Štěchová, Adam Olejár, Lucie Stroupková, and Edita Fedorková for assisting during data collection; and Jonathan Koppel for providing valuable feedback on an early draft of the manuscript.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 An immediate check after Session 1 confirmed that each participant worked on at least one distractor task from the selection.

2 Theoretically, participants could report and correctly attribute all words to a specific list and then recall some additional words from other lists, in which case internal intrusions would not have consequences on correct recall. However, reporting of more than 9 words per list was rare.

3 One reason for the weakness of the effect of the content deviation may be the size of the category that was changed (two words). Therefore, we decided to drop the category-size distinction in the replication Experiment 2.

4 Participants perceived the whole experiment as moderately difficult (from 1 = easy to 7 = difficult; M = 4.48, SD = 1.25) and had moderate motivation to complete each of the recall sessions (scale from 1 = not at all motivated to 7 = highly motivated; M = 4.81, SD = 1.45). These data were not used for any statistical analyses.

5 We arrived at this model by comparing three models: (i) fixed-effects-only, (ii) random intercepts for list nested within participants, and (iii) random slopes for delay across list nested within participants and found that the last one showed the best fit.

6 Looking back at our metacognitive measure of the awareness of the original organisation of the word-lists, we realised that participants often mentioned that the word-lists comprised words from specific categories, but less often explicitly mentioned the order. They usually expressed the order as an example, and if the example matched the original order of the categories in the lists, we scored it as correct. Using the correct sequence, however, does not necessarily represent awareness of the order as the organising principle.

7 Another way of looking at recall organisation is to ask participants whether they had used any recall strategies (because these are usually based on some level of organisation). Encoding and recall strategies of our participants typically matched – they tried to make use of the encoding strategy during recall. Approximately 70% of participants reported a recall strategy during encoding that was associated with images or stories that aimed to create connections among the words from a list. Two thirds of these participants mentioned selecting words from categories, which indicates that clustering might appear in recall. However, for the purposes of data analysis, we decided to measure categorisation (clustering) and sequencing based on recall protocols.