A Study On The Short-Term Memory Of Numerals In The Different Background Noise

Short-term memory had always been one of the main subjects which cognitive psychologists concerned about from 1950's to 1970's. With Baddeley & Hitch (1974) using working memory to refer to a system comprising multiple components, in contrast to short-term memory which emphasized the simple storage capacity, the term working memory has become increasing common and the concept of working memory has continued to be a fruitful one. Also, the last decades has seen the proliferation of research that combined attention process with memory phenomenon, and the so-called "irrelevant sound effect " is exactly one manifestation of this research trend. In the irrelevant sound paradigm, participants are shown a sequence of visual items (such as digits or consonants) to remember for about 10 s, after this retention interval they are asked to recall the items in the order in which they are presented. Some such trials are accompanied by irrelevant sound; others take place in quiet. Despite explicit instructions to ignore the sound, recall is markedly impaired by the irrelevant sound. As is known, noise is one of the severe environmental problems that faced modern people and has many negative effects to people's common life. Concurrently, because so much of work activity is now verbal and takes place in offices, the phenomenon of irrelevant sound experiments has much relevance for performance efficiency at the workplace.The present study provides an experimental test of numeric short-term memory span task and the classical Sternberg reaction task in different background noise. The background noise varied at three sound intensity levels (40.7, 56.4 and 73.3dBA) that have been classified as three types that are traffic noise, common life noise, and slow music. The results are following.(1) In different background noise, short-term memory span performance shows no significant difference, which still changes within 7±2 chunks.(2) In Sternberg reaction task, the change of noise intensity produces different interference on the general reaction time performance and the whole percentage which including outliers (extreme-reaction data) and errors (wrong-reaction data), and the difference mainly takes place between the low noise intensity and the medium noise intensity.(3) With the accretion of memory set, the reaction time of right-reaction that theoutliers have been discarded exhibits a linear rising movement. The reaction time in low noise intensity is significantly longer than the one in medium noise intensity. The difference between "Yes" reaction time and "no" reaction time at diverse memory set is not found, except that when the memory set is one, "Yes" reaction time is most significantly shorter than "no" reaction time.(4) The effect of changing noise intensity on performance is also found in the whole percentage that including outliers (extreme-reaction data) and errors (wrong-reaction data) at different memory set. With the increasing of memory set, all the three percentage results (outliers, errors, and both outliers and errors) display an unorderly changing trend.(5) There is also some difference among those short-term memory search speed in different background noise.