Dewar K M, Cuddy L L, Mewhort D J
J Exp Psychol Hum Learn. 1977 Jan;3(1):60-7.
Sequences of seven tones were presented, and recognition memory for individual tones of each sequence was tested under varying degrees of context. With no context, the test required recognition of a tone isolated from the sequence; with full context, the tone to be recognized was embedded in the original sequence. A series of three experiments demonstrated that recognition memory was far more accurate under full-context conditions than under no-context conditions and that the superiority was not wholly attributable to the serial position information or the order information provided by the full context. It is suggested that in addition to the processing of limited information for the pitch of isolated tones, pattern (or relational) information is abstracted from a tone sequence and is retrieved in the presence of full context.
呈现了七个音调的序列,并在不同程度的语境下测试了对每个序列中单个音调的识别记忆。在没有语境的情况下,测试要求识别从序列中分离出来的一个音调;在有完整语境的情况下,要识别的音调嵌入在原始序列中。一系列三个实验表明,在完整语境条件下的识别记忆比在没有语境条件下要准确得多,而且这种优势并非完全归因于完整语境提供的序列位置信息或顺序信息。研究表明,除了处理孤立音调音高的有限信息外,还从音调序列中提取了模式(或关系)信息,并在有完整语境的情况下进行检索。