Degree Name

Master of Arts (MA)

Semester of Degree Completion

1977

Thesis Director

John J. Rearden

Abstract

The experiment was designed to evaluate the role of position in serial learning. In the position learning task, subjects learned two serial lists presented alternately, then learned a subsequent paired associate list consisting of items of the previously learned serial list. The paired associate list contained three types of pairs; the same position-same pair, different position-same pair, and different position-different pair. The findings of the experiment supported the position hypothesis that if position is the cue for the recall of the item in a serial list, then, a subsequent paired associate list that is made up of the two previously learned serial lists will produce maximum positive transfer for the same position-same pair items. For the pairs that were constructed so that items occupying the same ordinal position in the serial lists were paired but were moved to a new position in the paired associate list, i.e., different position-same pair, the percentage of correct response was significantly less than the same position-same pair items and significantly greater than different position-different pair items; F=222.18, df=2. The latter pairs were constructed of syllables from the serial lists but were in a new ordinal position and were paired with a syllable that had a different ordinal position. The Tukey multiple comparison test for the differences between the three types of items was significant at ∝=.01; CR=1016.59. Furthermore, the linear trend in the data was striking and yielded an orthogonal contrast that was highly significant; F=443.83. The quadratic function was not significant, F=.53. The findings contradict both the remote association and sequential hypothesis in serial learning. The highly significant F for the different levels of position was F=222.1826, df=2 is reflected in Ω2=.34, which indicated that 34% of variation is accounted for by the different levels of position. The results provided evidence that it is not only same position items that caused positive transfer from serial lists to a paired associate list, but also the same paired items that are in a new ordinal position.

Download

Share

COinS