Eliciting children's recall of events: how do computers compare with humans?

This study investigated the usefulness of an interactive computer program in eliciting children's reports about an event. Fifty-four 5–6- and fifty-nine  7–8-year old children participated in an event with their regular class teacher which involved several activities and a mildly negative secret. Four days and again 14 days later, the children were interviewed individually by computer (alone) or by a human interviewer. The computer program incorporated animation and audio whereby an animated figure asked the questions and the children were required to provide a verbal response. The accuracy and detail of the children’s reports was similar across the interview conditions. The children were more willing to review their answers with the computer than the adult interviewer. However, responses to the computer were less consistent across the interviews, and the children were less willing to disclose the secret in the second interview to the computer compared with the human interviewer. Overall, the computer revealed little benefit in eliciting children’s recall of the event over the standard face-to-face interview.

Validity and reliability of a physical activity recall instrument among overweight and non-overweight men and women

This study examined the validity and reliability of survey items measuring one-week recall of physical activity (PA) and examined differences by weight status. A sub-aim of the study was to objectively assess the intensity of activity that most closely matches self-report. A questionnaire was administered to adults twice, three days apart. It was again administered after subjects wore a MTI/CSA accelerometer for seven days (n = 118). Several metabolic equivalent (MET) thresholds were applied to the accelerometer data. Agreement between test and re- test estimates of sufficient physical activity for health benefits (150 min/week) was high (% agreement > 90%). Correlations (rho) between total reported PA (mins/day) and accelerometer data were 0.29 (p < 0.05) among men and 0.25 (p < 0.05) among women. Among men, self- reported duration of moderate PA (3-5.9 METS) and accelerometer data were significantly correlated (rho = 0.40, p < 0.01), with no differences by weight status. Among women, a significant relationship was found only for those who were not overweight (rho = 0.52, p < 0.001). A significant correlation between self-reported duration of vigorous PA (6+ METS) and accelerometer data was only found for overweight men (rho = 0.40, p < 0.05). When lower MET thresholds were applied to the accelerometer data, women's reported duration of moderate-intensity PA was most strongly correlated with moderate PA (accelerometer) defined as 2.0-5.9 METS (rho = 0.39, p < 0.01). The recall instrument provides a consistent measure of physical activity and validation coefficients were similar to those obtained for other physical activity recall questionnaires. However, the ability to measure PA by self-report may vary by weight status.

Improving children's recall of an occurrence of a repeated event: is it a matter of helping them to generate options?

Three experiments were conducted to explore whether children's recall of an occurrence of a repeated event could be improved by encouraging them to consider various details that occurred across a series of events prior to making a judgement about which details were included in the target (to-be-recalled) occurrence. Experiment 1 explored whether children's recall of the target occurrence was better after the interviewer presented all the items from the series prior to the child identifying the final item. Experiment 2 explored whether having the children generate all the items facilitated their subsequent recall of the target occurrence. Finally, Experiment 3 directly compared the effectiveness of the above 2 procedures. Regardless of the children's age, the retention interval, or the type of item, children's capacity to identify which details were included in a target occurrence was enhanced when they were initially provided with all the possible details from the series of events. However, without relying on the interviewer to generate the options, the benefit of the technique was directly contingent on the children's ability to generate content details; this was a distinct source of difficulty for the children. Indeed, having children generate options had no beneficial effect on decisions about the temporal position of items unless performance was made conditional on the children's ability to remember the relevant details in the first place. The implications of the findings for the legal setting and for future research are discussed.

Do logbooks influence recall of physical activity in validation studies?

Purpose: To examine whether physical activity logbooks influence estimates of validity of 7-d recall physical activity questionnaires. Methods: A  convenience sample of 551 adults aged 18–75 yr wore an MTI  accelerometer for seven consecutive days and were then randomly administered two of four 7-d recall physical activity questionnaires that varied in length and format (Active Australia Survey (AAS), long and short  International Physical Activity Questionnaires (IPAQ-L and IPAQ-S), and Behavioral Risk Factor Surveillance System (BRFSS)). A subsample of 75% concurrently completed a physical activity logbook. Results: Correlations (rho) between self-reported and measured duration of moderate- and vigorous-intensity activity and total activity were similar among participants who received a logbook and those who did not for each of the four instruments. There was also no interaction between assessment method (survey, accelerometer) and the assignment of a logbook. For the IPAQ-L, however, variability in the difference between accelerometer data and responses to the vigorous items was smaller among those assigned a logbook (F = 4.128, df = 260, P = 0.043). Overall, there were no differences in percent agreement or kappa for participation in sufficient levels of physical activity according to receipt of a logbook for any of the surveys. Conclusion: The process of self-monitoring through completion of a logbook does not appear to influence estimates of validity for brief or long questionnaires with global questions. Whereas the magnitude of error in accuracy of recall of particular types of activity may be reduced by completion of a logbook that is similar in structure to the survey being validated, this does not appear to influence overall estimates of validity.

The effect of intellectual disability on children's recall of an event across different question types

This research examined the performance of 80 children aged 9–12 years with either a mild and moderate intellectual disability when recalling an innocuous event that was staged in their school. The children actively participated in a 30-min magic show, which included 21 specific target items. The first interview (held 3 days after the magic show) provided false and true biasing information about these 21 items. The second interview (held the following day) was designed to elicit the children's recall of the target details using the least number of specific prompts possible. The children's performance was compared with that of 2 control groups; a group of mainstream children matched for mental age and a group of mainstream children matched for chronological age. Overall, this study showed that children with either a mild or moderate intellectual disability can provide accurate and highly specific event-related information. However, their recall is less complete and less clear in response to free-narrative prompts and less accurate in response to specific questions when compared to both the mainstream age-matched groups. The implications of the findings for legal professionals and researchers are discussed.

The effect of events context on children's recall of non-experienced events across mutiple interviews

Purpose: The current study examined whether young children's willingness to assent to, and provide details about, a false (non-experienced) activity differs depending on whether the activity was allegedly embedded within (a) a specific event or (b) a broad (non-specified) time frame.
Method:  Ninety-nine children aged 4--5 years (from both low and high socioeconomic backgrounds) either (a) participated in a staged event that consisted of two activities or (b) did not participate in the staged event. One or two days later, all children were given false suggestions about a non-experienced (false) activity that had either high or low plausibility. Approximately 8, 15, and 22 days after the event, children were asked to recall the activities, and to answer a series of specific cued-recall questions.
Results: There was no effect of event context on assent rates, and the rate at which children reported interviewer suggestions. However, children who participated in the staged event provided fewer details about the false activity. Further, among those children who assented to the false activity, fewer subjects, objects, actions, temporal markers, locations, fantastic/improbable details, and confabulation errors were reported when the activity was embedded within the specific staged event.
Conclusion: The degree of error in children's accounts of a completely false activity is reduced when the activity is suggested to have occurred within a specified event as opposed to a broad (non-specified) time frame.

The inhibiting effect of brand salience on brand name recall

This paper reports on a replication of Alba and Chattopadhyay’s (1986) study of the effects of substantially heightened brand salience upon the recall of competing brand names. Heightened salience was consistently shown to have an inhibiting effect on recall across a variety of experimental conditions. However, in the replication study this salience effect was not observed. Instead a trend in the reverse direction was found. This new finding is congruent with associative network model of memory and its prediction that subjects concentrating on a brand should trigger links in memory to the brand and other brands in the category.

How well does the gudjonsson suggestibility scale for children, version 2 predict the recall of false details among children with and without intellectual disabilities?

Purpose. This study explored the effectiveness of the Gudjonsson Suggestibility Scale for Children, version 2 in predicting the tendency of older school-aged children (with and without intellectual disabilities) to generate errors in an independent suggestibility paradigm.

Method. Sixty-nine children with an intellectual disability (aged 9-14 years) and 50 mainstream children matched for chronological age participated in a 30-minute magic show that was staged at their school. Three days later, the children participated in a separate biasing interview that provided seven true and seven false details about the magic show. The following day, the children participated in a second interview where they were required to recall the magic show in their own words and answer a series of cued-recall questions. Between 1 and 2 weeks later, the children were administered the Gudjonsson Suggestibility Scale-2 (GSS-2).

Results. While there was no significant association between performance on the GSS-2 and the independent suggestibility paradigm for the children with an intellectual disability, the chronological age-matched children's yield scores predicted their reporting of both false-new details and false-interviewer suggestions for the independent event.

Conclusion. When predicting children's recall of false details, the GSS-2 appears to be more useful with mainstream school-aged children compared with children who have an intellectual disability.

Checks and cheques : implementing a population health and recall system to improve coverage of patients with diabetes in a rural general practice

Identification of all diabetic patients in the population is essential if diabetic care is to be effective in achieving the targets of the St Vincent Declaration.¹ The challenge therefore is to establish population based monitoring and control systems by means of state of the art technology in order to achieve quality assurance in the provision of care for patients with diabetes. ^2,3 Disease management receives extensive international support as the most appropriate approach to organising and delivering healthcare for chronic conditions like diabetes.⁴ This approach is achieved through a combination of guidelines for practice, patient education, consultations and follow up using a planned team approach and a strong focus on continuous quality improvement using information technology. ^5,6 The current software (Medical Director) could not easily meet these requirements which led us to adopt a trial of Ferret. In designing this project we used change management⁷ and the plan, do, study, act cycle⁸ illustrated in Diagram 1.