Structural Equation Models of Noise Effects onMemory and Attention

摘要

Three recent noise experiments employed the same noise conditions, design, procedure, and memory tests. Road traffic noise or irrelevant speech, presented at 66 dBA Leq, wererncompared to a quiet control condition (38 dBA Leq). The road traffic noise was made with a background of continuous traffic noise (～62 dBA) with superimposed segments of truck passages. The meaningful irrelevant speech recording consisted of background babble (～62 dBA), with superimposed segments from a conversation between teenagers. The test battery included a total of 15 different memory tests, yielding 18 performance measures, and was put together to reflect episodicrnmemory, semantic memory and attention. For memory tasks with a time differentiation between encoding and retrieval, retrieval was performed in silence. Participants were children aged 13-14rnyears, young adults aged 18-20 years, and teachers aged 35-45 and 55-65 years, making up a total of 288 participants. Structural equation modeling of the memory sub-systems showed attention, episodic, and semantic memory to be roughly invariant across noise exposure, age and gender. Thus, noise effects on memory do not operate by any allocation shift in the way cognitive resources are being employed. One important implication of this is that noise from irrelevant speech is not acted upon in any way different from noise from road traffic noise, and thus the cognitive strategies for handling these noise sources do not differ between themselves or from the quiet control group.