You are in: Home » Research at the Unit » Speech and Language
Short-term memory (STM)
Because spoken language unfolds over time, speech comprehension and production depend critically on our ability to store information for short periods of time. We need to store the results of any analysis of the immediate input so that we can integrate that input with subsequent information. Most importantly, we have to store that information in the correct serial order. "Man bites dog" does not have the same meaning as "dog bites man". "Dog" is not the same word as "god". Our ability to understand speech, and our ability to represent serial order in memory, are therefore inextricably linked. Memory for serial order also plays an important role in the acquisition of vocabulary. The individual components of the speech signal must be retained in memory if the infant is to appreciate that a particular sequence of sounds forms a word. As recent research has demonstrated, our ability to acquire vocabulary is critically dependent on phonological short-term memory (Baddeley, Gathercole & Papagno, 1998).
Spoken word perception, vocabulary acquisition, and short-term memory, therefore all share a common computational requirement. They must represent ordered phonological information in memory. The goal of this project is to increase our understanding of the relationship between the mechanisms responsible for the storage and representation of serial order in language and memory. The long-term aim is to develop a full account of how phonological information is stored in STM, how phonological representations of words can be learned, and how the short and long-term representations are combined in recognising continuous spoken input. A major component of this research involves development of a computational model of short-term memory. This part of the programme is designed to extend to scope of the Primacy model (Page and Norris, 1997) to explain the role of STM in learning long-term representations. Ultimately we would hope to be able to integrate this computational model of STM with the Shortlist (Norris, 1994) model of spoken word recognition.
Primacy Model (Page and Norris, 1997)
How is it that we can remember sentences or telephone numbers? More specifically, how can we remember the digits in a telephone number in the correct order? Although STM is one of the most extensively studied topics in cognitive psychology, we still don't have definitive answers to these questions. But, we have made considerable progress on this topic in recent years. Much of that progress is attributable to the development of new computational models. For many years, the most influential and productive theory of STM has been the Working Memory model of Baddeley and Hitch (1986). However, the Working Memory model only made qualitative predictions, and didn't provide a detailed explanation of how order could be represented in memory. To address this shortcoming, Mike Page and I have developed a computational model of STM called the Primacy model (Page and Norris, 1998). Based on quite simple assumptions, the Primacy model gives a detailed quantitative account of a wide range of empirical data.