THE EFFECT OF THE INSTRUCTIONS ON FACE RECOGNITION: ACCURACY AND EYE MOVEMENTS

"The present experiment examines how instructions (absolute judgement vs. relative judgement) affect the performance in simultaneous lineups (present perpetrator and absent perpetrator). To find out whether the participants really followed the instructions, their eye movements were recorded when they faced the photo lineup. Sixty participants (44 women and 16 men) took part in the experiment. Overall, the results showed that participants with absolute judgement instructions made significantly less inter-photograph comparisons than those with relative judgement instructions. In the present perpetrator lineup, hit rate was lower for participants with absolute judgement instructions than with relative judgement instructions. In the absent perpetrator lineup, no differences were between both instruction conditions. Furthermore, as was expected, no relationship was found between “pre” and “post” confidence and accuracy in the lineups. Moreover, we examined participants’ metamemory evaluations about their examination pattern of the photographs in the lineup. Our results did not show high incongruity between the own participants’ judgment and their visual behavior."

Information foraging for perceptual decisions

We tested an information foraging framework to characterise the mechanisms that drive active (visual) sampling behaviour in decision problems that involve multiple sources of information. Experiments 1-3 involved participants making an absolute judgement about the direction of motion of a single random dot motion pattern. In Experiment 4, participants made a relative comparison between two motion patterns that could only be sampled sequentially. Our results show that: (i) Information (about noisy motion information) grows to an asymptotic level that depends on the quality of the information source; (ii) The limited growth is due to unequal weighting of the incoming sensory evidence, with early samples being weighted more heavily; (iii) Little information is lost once a new source of information is being sampled; (iv) The point at which the observer switches from one source to another is governed by on-line monitoring of his or her degree of (un)certainty about the sampled source. These findings demonstrate that the sampling strategy in perceptual decision-making is under some direct control by ongoing cognitive processing. More specifically, participants are able to track a measure of (un)certainty and use this information to guide their sampling behaviour.

Multiple Monitors and Knowledge of Results in Vigilance: The Decrement Still Wins

Research has demonstrated that knowledge of results (KR) can improve vigilance performance and that certain types of KR can even eliminate the vigilance decrement. Likewise, researchers have shown that having a number of operators perform a vigil in a multiple-monitor setting can also increase performance on sustained attention tasks. The present study was conducted to determine if using KR in a multiple-monitor setting would eliminate the vigilance decrement and improve performance as measured by signal detection indices. Subjects completed an absolute judgement task either in pairs or individually and received KR to hits, false alarms, misses, or no KR. The results indicated that even KR in a multiple-monitor paradigm was not sufficient to combat the vigilance decrement. Characteristics of the task, the social interaction between monitors, the use of signal detection indices, and the practical implications of the findings are discussed.

Interactions in the Discrimination and Absolute Judgement of Orientation and Length

An asymmetric model is described for interactions in the perception of two dimensions (length and orientation) of a single visual stimulus. Two methods were used to test these interactions, and models for the interpretation of the possible outcomes of these tests are discussed. A length discrimination task showed facilitation (decreased reaction time) when orientation was covaried with length, and interference (increased reaction time) when random orientation variation was introduced. A smaller effect was seen when length was varied in an orientation discrimination task in a correlated or random fashion. Analysis of sequential effects showed that reaction times are fastest on repetition trials and are slowed by either the need to change the response or the need for additional sensory processing. With the second method, it was found that the amount of information transmitted in the estimation of orientation was not affected by the introduction of the redundant dimension of length, but that there was a significant gain in the amount of information transmitted in the estimation of length by the addition of the redundant dimension of orientation. It is concluded that orientation is probably a perceptual primitive of the visual system whereas length is a computed variable.

Geometry in Visual Space—Some Method-Dependent (Arti)Facts

The measure of scalar curvature K varies with the method used to determine it. Overconstancy is found only with simultaneous comparison, nearly perfect constancy results from successive judgement, while absolute judgement produces underconstancy.

The Complication Experiment Uncomplicated

Early work on the complication experiment, which involved the subject judging the position of a moving pointer at the onset of an auditory signal, is reviewed. The traditional explanation in terms of the earlier processing of input in one modality (the prior entry hypothesis) is critically discussed and rejected. Two experiments are reported in which constant error is shown to vary as a function of the position at which the auditory signal was delivered and the direction of rotation of the pointer. A third experiment found that auditory signals of different intensities gave rise to errors of similar magnitude. Unlike other recent complication experiments, all three gave rise to predominantly anticlockwise constant error. This disagreement was resolved by a fourth experiment in which predominantly clockwise error was obtained by presenting the auditory signal on only one revolution per trial. An explanation, developed from models of absolute judgement tasks, is put forward to account for the present and earlier results in terms of subjects’ judgement strategy.