Harry P. Bahrick

Maintenance of Foreign Language Vocabulary and the Spacing Effect

In a 9-year longitudinal investigation, 4 subjects learned and relearned 300 English-foreign language word pairs. Either 13 or 26 relearning sessions were administered at intervals of 14, 28, or 56 days. Retention was tested for 1.2.3. or 5 years after training terminated. The longer intersession intervals slowed down acquisition slightly, but this disadvantage during training was offset by substantially higher retention. Thirteen retraining sessions spaced at 56 days yielded retention comparable to 26 sessions spaced at 14 days. The retention benefit due to additional sessions was independent of the benefit due to spacing, and both variables facilitated retention of words regardless of difficulty level and of the consistency of retrieval during training. The benefits of spaced retrieval practice to long-term maintenance of access to academic knowledge areas are discussed.

Retention of Spanish vocabulary over 8 years.

Thirty-five individuals who had learned and relearaed 50 English-Spanish word pairs were tested for recall and recognition after an interval of 8 years. Two variables, the spacing between successive releaming sessions and the number of presentations required to encode individual word pairs, are excellent predictors of the likelihood of achieving permastore retention. Optimum recall occurs for words encoded in 1 -2 presentations and accessed at intervals of 30 days. Both variables yield monotonic retention functions that account for a range of variation from 0% to 23% recall. These variables also have very significant effects on the recognition of unrecalled words. A recent investigation of the retention of Spanish language learned in school (Bahrick, 1984) shows that a portion of the acquired knowledge has a life span of more than 25 years even if the knowledge is not rehearsed or accessed during that long interval. Another part of the originally acquired knowledge is lost within 5 years after training terminates, and virtually no knowledge is lost during the interval between 5 and 25 years following acquisition. This finding suggests the challenging possibilities of identifying conditions of learning and/or characteristics of material associated with a prospective life span of more than 25 years and of differentiating these conditions and characteristics from those associated with material destined to have a relatively short life span (less than 5 years). Such information would augment the very limited knowledge of life span memory currently available and make memory research more relevant to the needs of educators who have an obvious interest in prolonging the life span of transmitted knowledge. However, the research is arduous because it requires longitudinal investigations to extend substantially beyond the 5-year period during which material with a short life span is likely to be forgotten. The foregoing considerations led us to search for relevant longitudinal data, and we realized that such data might be available to us if we conducted a follow-up to an earlier investigation (Bahrick, 1979). The aim of the earlier investigation was to establish the effect of various rehearsal schedules on maintaining access to learned material over periods of 1 -9 months. One part of that investigation required college students to learn and relearn 50 English-Spanish word pairs in successive training sessions, spaced at intervals varying from a few seconds

Accuracy and Distortion in Memory for High School Grades

The relation between accuracy and distortion of autobiographical memory content was examined by verifying 3,220 high school grades recalled by 99 college students Accuracy of recall declined monotonically with letter grade, from 89% for grades of A to 29% for grades of D The positive correlation between achievement and accuracy of recall is attributed to more frequent rehearsals of affectively positive content and to greater accuracy of reconstructive inferences based on homogeneous, generic memories Most errors inflated the verified grade, and the degree of asymmetry of the error distribution is used as an index of the degree of distortion Distortions are attributed to reconstructions in a positive, emotionally gratifying direction Contrary to expectation, the percentage of accurate recall and the degree of asymmetry of the error distribution were uncorrelated This finding indicates that the process of distortion does not cause forgetting of the veridical content Rather, distortion reflects bias in reconstructive inferences that occur after the veridical content has been forgotten for other reasons

Lifetime maintenance of high school mathematics content

An analysis of life span memory identifies those variables that affect losses in recall and recognition of the content of high school algebra and geometry courses.

A re-examination of the interrelations among measures of retention

The interrelations among measures of anticipation, recognition and savings are examined. It is shown that (a) the difficulty level of a recognition task can be above or below that of anticipation for the same material, and (b) the slope of retention curves based upon recognition measures may be more or less steep than the slope of curves based upon recall measures of the same material. Previous contrary conclusions reflect the exclusive use of easy recognition tests, and experimental designs in which the degree of learning is much greater for the recognition than for the recall task.

Fifty years of language maintenance and language dominance in bilingual Hispanic immigrants.

Spanish language tests of 801 Cuban and Mexican immigrants showed no evidence of language loss during 50 years of U.S. residence; a few years after immigration, their English vocabulary approximated that of English monolinguals. The critical-age hypothesis was not supported for the acquisition of English vocabulary when English schooling and language usage were controlled by multiple regression. Most Ss continued to speak about as much Spanish as English; but read, wrote, and heard (on television and radio) far more English than Spanish. Under these conditions, Ss maintained Spanish dominance on tests of vocabulary recognition, lexical decision, and oral comprehension. Dominance was task specific and shifted to English on a category generation task about 12 years after immigration. No evidence of bilingual language interference was found; this is attributed to the strong Spanish foundation of the participants.

Stabilized memory of unrehearsed knowledge

The findings of Conway, Cohen, and Stanhope (1991) confirm that stabilized retention functions for unrehearsed academic knowledge are not an artifact of grade inflation as Hintzman (in press) has alleged. The Conway et al. data confirm that retention of unrehearsed knowledge may stabilize at the terminal level achieved during acquisition and that metacognitive confidence may decline during periods of stable retention. Low predictive power of grades for retention reported by Conway et al. is attributed to the nonhierarchival knowledge structures used in their investigation

The Cognitive Map of a City: Fifty Years of Learning and Memory

Publisher Summary This chapter presents a research that describes the acquisition of spatial information about a city during a four-year period of residence and the subsequent loss of this information over a period of 46 years. This is a study of long-term memory. The chapter presents the administration of five subtests, which yield 23 indicants of knowledge. The interrelations among these indicants provide a broad view of the available spatial information. The method of cross-sectional adjustment is used to compensate for the lack of control over the variables of acquisition and rehearsal. The acquisition data for most indicants show that more knowledge is gained during the first three weeks of residence than during the later periods of comparable length. After the first three weeks, street names and locations are learned at an even rate of about 2–3 streets per academic year without diminution at the end of four years. This rate is approximately the same for indicants of free recall, verbally or visually cued recall, or matching tests and includes the indicants of verbal and spatial knowledge. Intercorrelations among all indicants of performance are calculated at the five stages of learning. When performance is stabilized, these intercorrelations reflect primarily the nature of the knowledge and scoring system, but during the acquisition process, the correlations reflect individual differences in the speed of learning. The result is a family of contour retention curves reflecting the estimated effect of changes in one rehearsal variable on the particular indicant of retention.

Relearning and the measurement of retention.

A demonstration experiment is described in which saving scores are computed for the attainment of a series of successive criteria during relearning. It is shown that the traditional terminal criterion is far less reliable than intermediate criteria. Saving scores based upon an average of successive criteria are shown to increase the validity of measurement of retention of an entire learning process. This is equally applicable to the measurement of transfer effects where the principle of saving scores continues to be widely used.

Fifty Years of Memory of College Grades: Accuracy and Distortions

One to 54 years after graduating, 276 alumni correctly recalled 3,025 of 3,967 college grades. Omission errors increased with the retention interval, and better students made fewer errors. Accuracy of recall increased with confidence in recall. Eighty-one percent of commission errors inflated the actual grade. Distortions occur soon after graduation, remain constant during the retention interval, and are greater for better students and for courses students enjoyed most. Confidence in recall is unrelated to distortion. Courses that were not freely recalled, but had to be cued, were recalled less accurately and with less distortion. The data support a supplementary theory of memory distortion. The theory assumes that forgetting and distorting memory content are relatively independent processes, that relevant generic memories are used to fill in gaps after episodic memory fails, that systematic distortions affect autobiographical content that is emotionally and motivationally valenced, and that most individuals supplement with content that is emotionally more gratifying than the veridical content. The data conflict with dynamic displacement theories according to which screen memories actively block access to unpleasant veridical content.