Back to Cognitive Section Home Page

Provisional Timetable

Authors abstracts A to D Authors abstracts E to L
Authors abstracts M to P Authors abstracts Q to Z

A PET study is reported replicating a motor sequence learning study known normally to involve marked cerebellar activation. Activation levels of 6 dyslexic adults were compared with controls matched for age and IQ. Between-group differences arose specifically in the ipsilateral cerebellum, with rCBF increases for the dyslexic group showing only 10% of the control increases. These results confirm behavioural findings of cerebellar deficit in dyslexic groups. A causal chain is derived linking cerebellar problems in infancy, via early expressive and receptive speech, to the established difficulties in phonological processing. Supporting data relating to slower execution of articulatory gestures are also presented.

Background beliefs and evidence interpretation

Background beliefs and Evidence Interpretation Previous research into how people select evidence to help them decide between alternative hypotheses has been interpreted as demonstrating that they misunderstand the concept of diagnosticty in that participants tend to select evidence with explicit consequences for just one of a pair of alternative hypotheses. We argue that such single pieces of evidence may sensibly be interpreted using one's background beliefs and go on to present the results of three experiments designed to investigate how people interpret statistical evidence with explicit consequences for just one of two hypotheses. Experiments 1 and 3 demonstrate that people do use their background beliefs when interpreting single pieces of statistical evidence whilst the results of Experiment 2 suggest that they interpret such information with reference only to the hypothesis to which the evidence explicitly relates.

The Central Executive's role in visuo-spatial memory updating.

A visuo-spatial working memory updating task was devised which required individuals to recall the serial positions of the last 4 cells highlighted in a visual display. The number of cells highlighted in sequence varied from 4 to 10 (the participant was unaware of how many would be highlighted). Thus the task is analogous to Morris & Jones' (1990) consonant memory updating procedure. Concurrent random letter generation (assumed to utilise the resources of the central executive) selectively impaired performance on the updating component of the task. The implications for the role of the central executive in visuo-spatial tasks are discussed.

Writing as a knowledge-constituting process.

Current cognitive models of writing characterise expert writing as an explicit problem-solving process in which knowledge is actively transformed to satisfy rhetorical goals. Galbraith (1999) has recently proposed a dual process model in which this problem-solving process is complemented by a more spontaneous process in which knowledge is constituted during text production. This paper will present the results of an experiment investigating the extent to which new ideas develop in writing as a function of individual differences in self-monitoring and different types of planning. The results provide evidence of two different kinds of transformation of thought, the characteristics of which conform to the predictions of the dual-process model and contradict those of pure problem-solving models.

Negative priming and ageing

This study addresses the question whether target selection difficulty accounts for age-related changes in negative priming effect. Young and old adults performed a set of letter naming tasks where target selection difficulty was manipulated by varying color similarity or spatial proximity of target and distractor letters. Results indicated that the amount of negative priming increased with increasing difficulty of selection. Nevertheless, negative priming effects were equivalent in young and old adults. These results do not support the target selection difficulty hypothesis and are problematic for the widely accepted view of age-related deficit in attentional inhibition.

Misnaming the Snodgrass pictures: Which factors affect picture naming accuracy in aphasics?

Aphasic patients were presented with the set of Snodgrass and Vanderwart pictures to name. We investigated the effects of the following variables on naming accuracy: age-of-acquisition, word frequency, name agreement, image agreement, visual complexity, imageability, and the mean naming latencies to the pictures from normal speakers. Age-of-acquisition was the most consistent predictor of accuracy, with all but two patients showing an effect. Effects of word frequency were found only in patients whose overall level of naming accuracy were low. Naming accuracy showed no consistent relationship with detailed profiles of the comprehension and object processing abilities of each aphasic. It would appear that age-of-acquisition has a powerful effect on the availability of lexical entries.

Saccade Patterns in Serial Visual Search

The effect of planning instruction on childrenUs text searching strategies Performance of 11 years-old good and poor comprehenders was compared in a searching task (question answering) before and after planning instructions. We suggest that one of the causes of poor comprehendersU inefficient searching is their lower planning ability; as a consequence, they spend more time re-reading unimportant paragraphs and searching randomly, often forgetting what they are looking for. We also took additional measures of planning, verbal and spatial working memory and impulsivity, to see how these factors predict searching ability. The implications of these findings for improving children’s search strategies will be discussed.

Unconcious advertising retreival and positive influence on product liking

Two experiments investigated the relationship between unconscious memory for advertising and product liking judgments. Using the Process Dissociation Procedure (Jacoby, 1991), estimates of conscious and unconscious memory of advertising for canned drinks were calculated and liking ratings taken for the products. At test, adverts when previously seen, yielded significantly greater product liking ratings than when presented as new. Following the Mere Exposure Effect (Zajonc, 1980), liking increases were observed in conjunction with high estimates of unconscious, and low estimates of conscious memory of the advertising. Moreover, a significant positive correlation occurred between the amounts of liking change and unconscious memory.

Witch Won? Repetition, Homophone, and Phonological Priming of Colour Picture Naming

In a series of experiments we compared the priming of picture naming by the prior reading aloud of: (a) the object names (e.g., witch-WITCH); (b) homophones of the object names (e.g., which-WITCH); and (c) phonologically similar words, with either shared final overlap (e.g., slight-KNIGHT) or shared initial overlap (e.g., knife-KNIGHT). All experiments found homophone priming effects that were of equivalent magnitude to repetition priming, both for immediate and delayed priming. Immediate but not delayed phonological priming was found for rhyming words, with no reliable effects from words with shared initial overlap. These results show that the prior production of both lemma and word-form representations and word-form representations alone produce equivalent priming of object naming.

The elusiveness of negative priming

In the last 20 years research into Negative Priming, (the delay seen after a distractor prime becomes a probe target), has had two foci: i) What is occurring to cause this delay? ii) Is this delay a marker for individual differences in inhibitory mechanisms in both normal and clinical populations? Very little interaction between these two streams of research has occurred, but there is now considerable evidence from the first question, which affects the outcome of the second. Negative priming does not generalise from one task to another within individuals, and is only very weakly stable on retest of the same task.

Exogenous control of selective frequency listening

Two studies are reported in which words are recalled when in the same or a different olfactory context as encoding. In Study 1, in which the words to be remembered were unrelated to the smell manipulation, a marginal effect of context-reinstatement was observed. In Study 2, a far stronger effect of context reinstatement was observed, but was largely limited to words that were ontologically related to the olfactory context. Subsequent semantic cueing revealed that smell-unrelated words could indeed be retrieved, indicating that smell, while a very effective cue for some materials, is not a very effective retrieval cue for others. © Crown copyright

Colin Hamilton

The Corsi Block tapping task has been employed through three decades of research as a tool for investigating visuo-spatial efficacy. This study examines the demands the task makes upon Visuo-Spatial Working Memory (VSWM). Performance in two patients across a range of working memory task procedures is investigated. The working memory tasks were constructed in order to place differential demands upon the slave and executive processes in VSWM. The patterns of results suggest an interesting dissociation between the VSWM tasks and Corsi Block task performance. The implications for the interpretation of Corsi task performance are discussed.

Thinking counterfactually about actions and inactions

The ability to consider how reality might have turned out differently has been implicated in a range of cognitive abilities involving causal and hypothetical reasoning. One result of how we generate counterfactual alternatives is that we tend to regret our actions rather than our failures to act even when the action and the inaction produce the same outcome. This tendency is known as the action effect in counterfactual reasoning. We report the results of a series of experiments demonstrating that both the order in which the action and inaction are described as well as the implications of a failure to act are variables moderating the effect. The consequences of these demonstrations for theories of counterfactual reasoning will be examined.

Connectionist modelling of conceptual knowledge development: A feature based approach

Children's knowledge acquisition processes and their relation to adult processing are poorly understood. Connectionist modelling could address this, however, little developmental data appropriate for network training and evaluation exists. To provide data, children completed category fluency and feature generation tasks. I will discuss how analysis illustrated children's principle bases for organisation of biological kinds, and describe a connectionist model being trained using the featural data collected. Comparing knowledge organisation in the model to that in children will shed light on knowledge acquisition processes, and enable evaluation of the ability of feature based theories to account for acquisition processes.

Communicating Music with Drawings: Representation in Graphical Communication

This paper reports an experimental study of cross-modal representation: the graphical communication of music. One subject produces a drawing of a target excerpt of piano music and their partner must decide which of two pieces correspond to the drawing. The only means of communication is via a shared, virtual whiteboard, and alphanumeric characters are excluded. The results show that although there are almost no established representational conventions that subjects can draw on for this task, they perform consistently above chance. Furthermore, the results suggest significant parallels between the representational strategies engaged by this task and those previously established for conversation.

Implicit and explicit preference decisions in artificial grammar learning

In two experiments, participants carried out an artificial grammar learning task, making either explicit or implicit preference decisions at test.. Increased exposure to learning strings resulted in better discrimination of novel grammatical from ungrammatical strings for explicit preference participants. The discrimination performance of implicit preference participants was unaffected. A secondary task at test did not affect the performance of implicit preference participants relative to single task conditions. Explicit preference performance was completely eliminated. It is concluded that explicit processing is sensitive to on-line interference, and to changes in the knowledge base upon which it acts. Implicit processing is relatively insensitive to such factors.

Spared and impaired memory following selective hippocampal lesions in humans.

Aggleton and Brown [1] argue that hippocampal lesions disrupt all forms of recollection, but spare item recognition when this is dependent only on familiarity. A neural network model [2] argues that, when targets and foils are very similar, yes/no item recognition taps recollection. An impairment on this task, but not forced-choice item recognition, is predicted following hippocampal lesions. We report a patient with selective bilateral hippocampal damage (YR). Consistent with the prediction, when test discriminations are difficult, yes/no item recognition is impaired but forced-choice item recognition is totally normal even though for control subjects the yes/no paradigm is the easier task. 1. Aggleton, J.P. and Brown, M. (in press) Behavioral and Brain Sciences 2. O'Reilly, R.C. and Rudy, J.W. (1999) Technical Report. University of Colorado, 99-01, 1-46.

Why do young speakers experience difficulty on function words?

For people who stutter (PWS), dysfluencies on function words mainly occur when bthey precede content words where the plan for the latter is not ready for execution. Repetition on function words gains planning time for the content word. Spontaneous speech data were drawn from five age groups. For fluent speakers and young PWS, more dysfluencies occurred on initial function words than content words. For other PWS, dysfluencies on function words decreased and that on content words increased. These suggest that fluent speakers use the delaying tactic whereas PWS attempt a content word with an incomplete plan which resulted in stuttering.

Separable speech rate effects on the development of verbal span.

A study is reported examining the association between speech rate and verbal short-term memory span in children. Verbal spans of 60 typically developing children, aged between 5 and 10 years, were assessed. Individuals speech rates were separately determined by timing repetitions of the words used in memory trials. This measure was then divided into the time spent articulating words, and the time spent pausing between words. Correlational analyses showed that both word and pause duration accounted for significant and independent variance in memory span. The implication of these findings for theories of verbal short-term memory development are discussed.

Metamemory, Ageing and the Frontal Lobes

This research examined the effect of age on confidence-accuracy relationships. In two experiments, young and older adults viewed a short film for which their memory was subsequently tested. In both experiments older adults were tested after a delay of 10 minutes. Younger adults were tested either after a delay of 10 minutes (exp 1) or, to match them on memory performance with the old, one hour (exp 2). The elderly exhibited impaired metamemory in both experiments, even when they were matched on recall performance with the young (exp 2). Importantly and additionaly, for the old, performance on tests of frontal lobe functioning correlated with metamemory but not memory performance.

The relationship between phonological working memory and vocabulary acquisition

The relationship between phonological working memory and vocabulary acquisition in poor readers was examined in order to determine whether they have difficulties in establishing phonological representations in long-term memory. A cross-sectional study of ten-year-old severely impaired poor readers showed deficits for reading age in the acquisition of new spoken words, measured by the ability to attach verbal labels to visual stimuli. It was also found that the poor readers showed significantly lower long-term recall, in delayed memory tasks, for both familiar and unfamiliar spoken words. However they had no impairment, for reading age, on phonological working memory measures (such as digit span and nonword repetition). It was concluded that the poor readers had severe problems in setting up phonological representations in long term memory, which could not be accounted for in terms of phonological working memory difficulties.

Lateralized Word Recognition: Assessing the Parallel-Sequential Distinction.

A popular view is that words are processed more efficiently in the right (RVF) than in the left (LVF) visual hemifield because of parallel versus sequential orthographic analyses. We investigated this view using the Reicher-Wheeler task to suppress effects of partial word information and an eye-tracker to ensure central fixations. RVF advantages for words obtained across all serial positions and "U-shaped" serial-position curves obtained for both visual hemifields. Moreover, whereas words and nonwords produced similar serial-position effects in each hemifield, only RVF stimuli produced a word-nonword effect. These findings support the view that left hemisphere function underlies the RVF advantage but not that different modes of orthographic analysis are used by each hemisphere.

Who needs nouns when verbs will do just as well?

Participants' eye movements were recorded as they inspected visual scenes whilst hearing sentences like 'the boy will move the cake' or 'the boy will eat the cake'. The cake was the only edible object in the scene. Eye movements to the cake were initiated after the onset of the spoken word 'cake' in the move condition, but before its onset in the 'eat' condition, even when participants were instructed to ignore each sentence. Information extracted at verbs can thus, in combination with contextual information, identify the entities that nouns ordinarily refer to. These data illustrate the predictive nature of sentence processing.

The developmental neuroscience of language: The case of Williams Syndrome.

Two experiments examine performance in a sequence learning task. Participants are trained on a repeating sequence which is presented as a visual display and learning is measured via the increase in RT to respond to a new sequence. Some participants make a response to each stimulus while others merely observe the sequence. In Experiment 1 participants responding to the display via a keypress show learning, but those merely observing do not. Five possible reasons for the failure to find observational learning are considered and the final experiment attempts to resolve these.

A golden or Bermuda triangle? The relationship between semantic impairment and surface dyslexia.

We examined the combined effects of plane and depth rotation on the efficiency of object naming. Subjects practised identifying only plane-rotated or only depth-rotated views of familiar objects. Initially, both plane-disoriented and foreshortened views of familiar objects were identified more slowly and less accurately than canonical views. Following practise, plane and depth rotation effects attenuated, but were not eliminated. Most importantly, practise at identifying stimuli rotated either in plane or in depth reduced rotation effects for unseen views in the untrained dimension (depth or plane rotation respectively) as well as for unseen views in the trained dimension (plane and depth respectively).

Optimum dimensionality and word co-occurrence statistics

Several recent papers describe how some lexical properties can be captured by simple measurements of which other words tend to occur close to the target word. The resulting co-occurrence statistics vectors have been used to account for phenomena ranging from semantic priming to vocabulary acquisition. Landauer and Dumais (1997) stress that to be of explanatory value the dimensionality of the co-occurrence vectors must be reduced to around 300 using a procedure related to principal components analysis. We show that by changing the methodology used to measure these statistics we can surpass their results without employing dimensionality reduction.

Processing and storage in visuo-spatial working memory

Intervention studies are presented which illustrate the advantages of flexible computer based learning (RITA) to support reading instruction. Study 1 found dramatic improvements in decoding and comprehension abilities (+5.35 and +7.9 months from a 10 week intervention) in eight, 11 year old children previously resistant to support (average RA-CA discrepancy at start -4.34 years) This illustrates the importance of combining reading comprehension and decoding instruction and the facilitative effect of RITA. Study 2 concentrated on comprehension, comparing reciprocal teaching and vocabulary enrichment. The advantages of using individualised programs of instruction based on previously identified subskill deficits are discussed.

Poster Presentations

Back to Cognitive Section Home Page

Provisional Timetable

Authors abstracts A to D Authors abstracts E to L
Authors abstracts M to P Authors abstracts Q to Z

This site is maintained by Dr. Geoff Ward, University of Essex (e-mail: gdward@essex.ac.uk). The Cognitive Section of the BPS would like to show the most useful and up-to-date information related to cognitive psychology. Please feel free to suggest any additional information, links, or corrections which you think may be useful.

---

Last modified 5/8/99