Time Processing in Human Behaviour and Evolution
Jill Boucher, University of Sheffield

It is generally agreed that what distinguishes people from chimps includes a certain type of social intelligence; language and other symbol systems; generativity not only in the use of symbol systems but also in technology, culture and visual art; and higher forms of consciousness. There is no agreement, however, as to how these capacities relate to each other in modern man, let alone how they related to each other during evolution. I argue that what made all of these things possible included - (I'm not claiming an exclusive role) - changes in the brain's ability to process time.
Animals and people have explicit biologically-based as well as implicit biologically-based brain mechanisms for processing time. Explicit time signals are produced by nerve cells in the brain which have regular on-off patterns of firing which provide oscillatory activity at parallel, harmonically-related frequencies or distributions of frequencies (Gallistel, 1990; Treisman et al., 1994; Brown et al., 1997). These mechanisms are involved in segmenting, subitising, and hierarchical organisation of temporally ordered perceptual information, in the organisation of temporally ordered motor outputs, and in the co-ordination of perception and action. It has recently been demonstrated that explicit time-marking mechanisms of the kind provided by oscillatory brain activity provide a temporal context in which to encode and retrieve information in short term memory (Burgess and Hitch, 1996; Brown et al., 1997). It follows that similar mechanisms are almost certainly involved in long term, episodic memory.
I argue that extensions to hominid oscillatory brain activity enabled homo sapiens to subitise incoming experience at an extended range of time chunks, simultaneously increasing the extent and complexity of the hierarchical time frame within which items can be encoded at input and organised at output, thus eliminating reliance on linear, associative chaining as an organisational principle for perception and action over extended time periods. This allows a move from assoicative and procedural learning to script learning, and a related move from the acquisition of categories to the acquisition of concepts. Put another way, items encoded within a hierarchical as opposed to a linear framework become context-free, and capable of what Karmiloff-Smith (1992) refers to as the first stage of representational redescription. Items which are abstract and context-free can be recombined, organised and run off-line utilising the extended temporal framework, resulting in generative thinking of the kind required for other-simulation, for extended planning, hypothetical thinking, and creativity. These advances in cognition, including social cognition, would have led to developments in tool-making and social organisation such as would have greatly increased pressure on communication. Protolanguage would have co-evolved with conceptual and script-based thought. Whether or not enhanced temporal processing (with appropriate computational capacities) can give us grammar is an open question, though I suspect the answer may be 'Yes'. Language gives us an additional layer of consciousness.

Memories, Remnants and Traces
David Cockburn, University of Wales at Lampeter

On a familiar philosophical model, to remember an event is to have a memory impression on the basis of which one concludes that what is represented in the impression happened in the past. This model leads to serious difficulties. We need to recognize that memory is not a source of knowledge of the past: one who remembers what happened is one who already knows what happened. Connected with this, we need to reject the claim the idea of a sharp contrast between a 'purely cognitive element' in remembering and an emotional response to the past event which is based on that. We need to place the idea of a concerned response to something in the past at the centre of our understanding of our relation to the past in memory.
We can place besides this other ways in which we are linked to the past. We are linked to lives in the more distant past through what is left of those lives in the present: their buildings, rituals, writings, and so on. I see the lives of people in such remnants. We can draw a contrast between treating, say, an ancient building as a 'remnant' and as a 'trace' of past lives: as the very building in which they lived and worked, or as simply an effect in the present from which we may draw conclusions about their lives. That contrast has application in other areas: for example, in relation to another's description of some event she witnessed. We may treat what she says as a sharing of knowledge that she possesses, or simply as an effect of a past event.
Returning to memory, while we need not agree that all remembering involves traces it is clear that the idea of a 'trace' provides a possible technique by which one may distance oneself from past events when the memory of them is too painful. Similarly, one may think of, say, one's shamed response to events of a certain class as simply an effect (a trace) of the past event, rather than as embodying an understanding of what happened. While there can, in particular cases, be good reasons for doing this, the philosophical model with which I opened presupposes that this is the only correct way to view one's relation to, and so one's understanding of, a past event.

The Self-Memory System and its Neuroanatomical Basis: Some Thoughts on Consciousness, the Self, and Autobiographical Memory
Martin Conway, University of Bristol

I will describe a model of autobiographical memory in which memories are transitory mental constructions within a self-memory system (SMS). The SMS contains an autobiographical knowledge base and current goals of the working self. Memories are constructed in the SMS by executive processes which control access to the knowledge base. Data from an Event Related Potentials study show networks in the frontal lobes mediate memory construction in right temporo-occipital areas. It is proposed that frontal activity reflects the operation of control processes and goals of the self whereas activity in temporal and sensory processing areas reflects activation in the knowledge base. This neuroanatomical circuit is compared to a 'consciousness' circuit originally proposed by Dimond (1977). Various aspects of consciousness in remembering are then considered from both neurobiological and cognitive perspectives.

Memory Processes Underlying Humans' Chronological Sense of the Past
William Friedman, Oberlin College, Ohio

An examination of past research shows that there are two distinct ways in which we remember the times of past events: as locations in time patterns and as distances from the present. This seminar is a review of what has been learned about the psychological processes underlying these two kinds of memory for time and the roles they play in cognition.

Egocentric and Objective Time
Robin Le Poidevin, Department of Philosophy, University of Leeds

Gareth Evans, John Campbell and others have articulated and discussed a distinction between 'egocentric' and 'non-egocentric' (or 'objective') representations of space. This is, roughly, the distinction between representations which do, and those which do not, include reference to one's own position. The extension of the egocentric/objective distinction to time would seem a natural one, but there are number of difficulties raised by such an extension, to do with important disanalogies which many philosophers have drawn between 'here' and 'now'. The aim of this paper is to address these difficulties, remove some confusions which surround the notions of egocentric and objective time, and attempt to show that considerations concerning memory provide reasons to accept a certain conception of objective time.

How are Unaware and Aware Memory Related?
Andrew Mayes, University of Sheffield

Explicit memory and item-specific implicit memory are generally supposed to differ primarily with respect to the fact that the former involves a feeling of awareness that a representation is of something encountered in some way in one's past whereas this is not true of the latter. A framework is proposed that comprises four kinds of mechanism: (1) The kind of information held in a memory representation and the brain processes required for the storage of this information, (2) the enhanced fluency of activation of a representation triggered by the encoding of its components, which results from the storage changes mediating memory, (3) the kind of attribution process that is triggered by enhanced fluency and can give rise to aware memory feelings as well as specific perceptual and aesthetic feelings, and (4) active search, which is an often present feature of explicit memory, but may never be required for implicit memory retrieval. It will be argued that item-specific implicit memory and explicit memory which are for the same information probably depend on the same memory representation and neural system, that item-specific implicit memory is equivalent to enhanced fluency, and that explicit or aware memory arises because enhanced fluency in certain contexts leads to a feeling of memory. Evidence supporting this argument will be considered.

Understanding the Past Tense
Christopher Peacocke, Magdalen College, Oxford

This presentation will concentrate on constitutive, philosophical issues concerning thought about the past. My aim is to identify the nature of the truth-conditions of statements and thoughts about the past; to suggest a constitutive theory of our understanding or grasp of the past tense which respects the nature of those truth-conditions; and to draw some lessons from the positive account I will be presenting. I will be touching on such issues as the role of truth-value links in understanding; the role of memory in temporal thought; and the nature of a realistic conception of the world and understanding. An effort will be made to keep the talk intelligible to an interdisciplinary audience.

Models of psychological time
J.H. Wearden, Department of Psychology, Manchester University

Contemporary timing theory is largely concerned with understanding sensitivity to duration. Two general classes of models for this process might be identified. One type uses quantity encoding; that is, the amount of time passed is reflected in the quantity of something which accumulates during the time period. The other class involves quality encoding, where different amounts of time are registered by the activation of different elements, not necessarily more or fewer. The dominant current model is that of an internal clock, consisting of a pacemaker connected to an accumulator, and examples of how this mechanism can represent time will be presented. Variants of the internal clock model include a "cognitive counter" which registers aspects of experienced events rather than clock "ticks". Such clock-like models are powerful and flexible tools for accounting for timing behaviour, but their very flexibility can raise problems of testability. An alternative to the clock model is the connectionist model of Church and Broadbent (1990), which embodies a form of quality encoding of time using an oscillator-based time representation. Another approach, which seems at first sight very different from the clock model, is the Behavioural Theory of Timing, where organisms are proposed to use their own overt behaviour to mediate time judgements rather than internal processes. Models which have no internal clock per se, such as those based on memory decay or spreading activation, will be discussed. Although some of these have potential for simple time representations, it is at present unclear whether they can ever be developed to account for a full range of timing behaviours. A recent focus of interest has been on how time might be represented in allegedly neurologically-plausible mechanisms, and the particular difficulties of doing this will also be discussed.