Cognitive Processing PDF

STAGES OF COGNITIVE PROCESSING GESTALT AND SCHEMA THEORIES Another of the author’s action slips involved absentmindedly adding instant coffee to a mug which already contained a teabag, thus creating a rather unpalatable hybrid beverage. Action slips of this kind have been extensively do...

STAGES OF COGNITIVE PROCESSING GESTALT AND SCHEMA THEORIES Another of the author’s action slips involved absentmindedly adding instant coffee to a mug which already contained a teabag, thus creating a rather unpalatable hybrid beverage. Action slips of this kind have been extensively documented and in most cases can be explained by the activation or perseveration of automatic processes which are not appropriate automatic processes can provide adequate control of our neural functions in most routine situations without needing to use up our attention, but they must be overridden by the conscious supervisory attention system when more complex or novel tasks require the flexibility of conscious control Automatic processes are obviously of great value to us, as they allow us to carry out routine tasks rapidly and without using up our limited attentional capacity. However, automatic processes lack flexibility, and when they fail to provide appropriate behaviour they need to be overridden by consciously controlled processing. There is some evidence that this override system may be located in the frontal lobes of the brain, since patients with frontal lesions are often found to exhibit perseveration of automatic behaviour and a  Well, one way to recognize your grandmother would be to have an internal schema or ‘template’ that could be compared with incoming sensory information. THE GESTALT APPROACH A key issue addressed by the Gestalt psychologists was the way that we might segregate the world into figures and the background against which they appear. This may sound trivial but is crucial as, if we are to recognise objects, we need to be able to tell them apart from everything else. The importance of figure and ground can be illustrated by one of the well-known reversible figures Perhaps one of the nicest conceptualisations of the feature- extraction approach was Pandemonium model which is illustrated in The way that the model works is that there are layers of ‘demons’. Demons at the lowest level in the system (remember this is essentially a bottom-up approach) PARALLEL DISTRIBUTED PROCESSING APPROACHES One way of getting around the problem of needing an almost infinite number of ‘grandmother cells’ in the brain is provided by parallel distributed processing (PDP) models PDP models are also sometimes referred to as connectionist or neural network models and these models, when implemented on computers, attempt to model the way in which the brain may work. In some ways, PDP approaches are still template approaches but the templates are much more flexible and, given that they represent stored knowledge, they are another conceptualisation of the schema VISUAL ILLUSIONS When something does go wrong with our perceptual system and what we perceive does not truly represent the outside world we usually refer to this as an illusion, and the study of such illusions provides valuable insights into how perception operates at different levels. Sensation will be considered to be the ‘raw’ bottom-up input from the senses and perception will be considered to be the end result of the processing of that sensory material within the visual system. The individual may be consciously aware of the perception arising from incoming sensory information, or they may not (subliminal perception). Sensation and perception thus lie at opposite ends of the visual process and may well be quite different. objects or events that exist independently of the senses as numena and our experience of those objects and events as phenomena. Kant argued that we can never truly access the numena, only the phenomena. That is, we can never know the world as it truly is, only our perception of it after it has been filtered and modified by our senses and cognitive processes. ‘We see things not as they are, but as we are.’ The term used to describe how easily an object can be detected by the senses is sensory conspicuity, and refers to the intrinsic properties of an object (such as shape, colour, brightness, amount of noise that it is making) that are likely to be registered by the senses – usually as a result of increasing contrast with the background. Thus, a pedestrian can often increase their sensory conspicuity by carrying a torch, or wearing reflective material  sensory conspicuity, although necessary for the detection of objects, may not always be sufficient for an individual to become aware of that object and take action to avoid it. in order to be able to consciously perceive (and react to) an object, it should also have high attention conspicuity. This term refers to the fact that to perceive something, the individual’s senses need to detect THE CONSTRUCTIVIST APPROACH: PERCEPTION FOR RECOGNITION One of the theories of the way in which perception operates and which deals explicitly with why we make so much use of stored knowledge is the constructivist theory. It is called a constructivist theory because it is based on the notion that it is necessary for us to ‘construct’ our perception of what we see from incomplete sensory information. Thus we use what we already know to fill in the gaps and interpret the sensory information we act as ‘scientists’, generating perceptual hypotheses (predictions) about what we may be seeing and testing those hypotheses against the sensory information coming in. The cat picture used previously can be used again to give an idea of how this works. The picture is not at all clear and may be difficult, at first, to resolve into anything that makes sense. You might thus generate a range of hypotheses about EVIDENCE FOR THE CONSTRUCTIVIST APPROACH: MASKING AND RE-ENTRANT PROCESSING demonstrated that changing one stimulus rapidly for another disrupted processing of the first stimulus, a process referred to as masking.. In a masking paradigm, a second stimulus can prevent recognition of an earlier stimulus if the mask follows very soon after presentation of the stimulus. It is not even necessary for the stimulus and mask to be at the same position in the visual field (i.e. not spatially coincident). A mask that  suggest that the mask is effective because it disrupts re- entrant processing. This term is used to describe the finding in neuroscience research that communication between different areas of the brain is never in one direction only. If a signal goes from one area to another, then there is sure to be one coming back the other way THE STRUCTURE OF THE VISUAL SYSTEM THE DORSAL AND VENTRAL STREAMS 1. The ventral stream is primarily concerned with recognition and identification of visual input whereas the dorsal stream provides information to drive visually guided behaviour such as pointing, grasping, etc. 2. The ventral system is better at processing fine detail whereas the dorsal system is better at processing motion although the differences are only relative 3. The ventral system appears to be knowledge-based using stored representations to recognise objects whilst the dorsal system appears to have only very short-term storage available 4. The dorsal system is faster 5. We appear to be more conscious of ventral stream functioning than dorsal. For instance individuals may report awareness of ventral processing, while manifesting different dorsal processing. A good example of this is if people actually interact with visual illusions, such as the hollow-face illusion. The perception is illusory, but the action (e.g. flicking a fly off the nose of the hollow face) does not appear to be influenced by the illusion. 6. The ventral system aims to recognise and identify objects and is thus object-centred. The dorsal system is driving some action in relation to an object and thus uses a viewer-centred frame of reference These characteristics support earlier research which suggested that the ventral stream is concerned with the question, ‘What is it?’ whereas the dorsal stream is concerned with the question, ‘Where is it?’ Thus, the ventral pathway is often known as a ‘what’ system, and the dorsal pathway a ‘where’ system it is suggested that the dorsal and ventral streams act synergistically, with the dorsal stream largely concerned with perception for action, and the ventral stream with perception for recognition. THE INTERACTION OF THE DORSAL AND VENTRAL STREAMS: PERCEPTION FOR RECOGNITION AND ACTION It is interesting to speculate, as we finish this discussion of visual perception, just how these two types of processing may act together to allow us to perceive our world. To do this, it is worth considering our experience and consciousness of what we are perceiving, i.e. our phenomenological experience. AUDITORY PERCEPTION AUDITORY LOCALISATION 1. Azimuth (horizontal), determined primarily by binaural cues, specifically time and intensity differences between stimuli reaching the left and right ears. Interaural intensity differences are largely due to the shadowing effect of the head that keeps high-frequency sounds from reaching the far ear. Long wavelengths (low-frequency sounds) are unaffected by the head, but shorter wavelengths (high-frequency sounds) are reflected back. This feature has been shown to be surprisingly useful in an evolutionary perspective. As a general rule, animals with 2. Elevation (vertical), determined mainly by spectral cues which are generated by the way in which the head and outer ears (pinnae) affect the frequencies in the stimulus. Sound reflected from the pinnae can be used to give an idea of the elevation of a sound. Thus the pinnae play an active role in sound localisation, suggesting that they did not evolve solely to rest spectacles on. If you fill in the pinnae with modelling clay (do not try this at home, although if you give the modelling clay to the two-year old that you gave the hammer to in , they will probably stick it in their ears without being asked; modelling clay is also likely to afford its use), sound localisation progressively worsens 3. Distance coordinate (how far a sound source is from the listener). Generally, judgements of distance for sounds within an arm’s length are good (interaural level difference (ILD) is large), but as sounds get further away, distance judgement is much more difficult, and distance to far-away sounds is generally underestimated. There are several mechanisms for auditory far-distance judgement, which are used together to determine perception of a sound’s distance Motion parallax –. Nearby sounds appear to shift location faster than sounds that are further away. This is analogous to the visual depth cue of motion parallax. TOP-DOWN INFLUENCES ON AUDITORY PERCEPTION A listener’s experience and frame of mind can influence how a message is perceived. demonstrated, with phantom word illusions, that people often report ‘hearing words related to what is on their minds’. The demonstration used simple words, e.g. ‘Boris’, ‘Go back’,‘Harvey’ played repeatedly and continuously. Listeners reported hearing new words and phrases that were not present in the original recording. As with vision, there is an interaction of top-down and bottom-up processing PROPRIOCEPTION, KINESTHESIS AND HAPTIC INFORMATION The sense that keeps track of the position of our body, limbs, fingers, etc. is known as proprioception and it operates through a system of nerve cell receptors (known as proprioceptors) that allow us to ascertain the angle of our various joints. A related sense, known as kinesthesis, allows us to discern how our body and limbs are moving and is a key element in such things as hand–eye coordination, and as such is a sense that can be improved through training and Attention WHAT IS ATTENTION? Attention refers to systems involved in the selection and prioritisation of information processing, and it is intimately linked with perception and memory and is thus central to almost everything we do WHAT IS ATTENTION FOR? Here I shall introduce many of the functions of attention to be addressed in detail later. An important question to ask is ‘what is attention for?’. identified two possible functions of selectivity in visual attention. First, there is selection for perception (i.e. detecting and selecting what to process from a visual display) and second, selection for action (i.e. detecting and selecting which response or action to make). However, in order to correctly combine what something is with where it is, these two sources of information must be correctly linked together. For example, if there is a red colour in the shape of a circle on the left, and a green colour in the shape of a square on the right, the colour, shape and position of each property must be correctly bound together. This is called the binding problem, and an important function of attention is to bind together what an object is, together with where it is and how to act on it. When attention is driven by our intentions it is called controlled attention, or executive control, and is said to operate top- down because it is influenced by a goal we have set ourselves such as searching for something in particular. The source of control is endogenous, it comes from within us. ‘Control is the set of fluid operations that enable intentional processing and adaptive performance’. In contrast, exogenous attention is stimulus- driven, where incoming or bottom-up stimuli trigger automatic processing which cannot be controlled intentionally. Patients with frontal lobe damage have difficulty inhibiting automatic responses and switching between tasks, as their mechanisms of attention appear to lack control. Theories of attentional control , aim to explain goal-directed behaviour and everyday slips of action. WHERE IS THE LIMIT? THE SEARCH FOR THE BOTTLENECK When two stimuli are presented in rapid succession and the participant must make a fast response to both, response time (RT) to the second stimulus depends upon the time interval between the presentation of the two stimuli At short inter-stimulus intervals, RT to the second stimulus is slower than when there is a longer interval.delay in responding to the second stimulus the psychological refractory period (PRP). the PRP demonstrated that there was a bottleneck in processing, so that at very presenting two different messages at once over headphones, one message to the left ear and the other message to the right ear. The participant was instructed to attend to one of the messages and repeat it back as it arrived, a procedure known as shadow participants were able to use ear location (left or right) and voice quality (male or female) to select the message to be attended. However, at the end of the experiment, when participants were quizzed on the meaning of the unattended message they were unable to say what it had been about. According to Broadbent, the unselected stimuli must wait in a highcapacity, fast-decay sensory memory, and unless they are selected for transfer before decaying they will be lost. So, if unselected stimuli do not get passed on to the identification stage they remain unattended and nothing is known about their meaning. This is called an early selection theory of attention, because selection of information for further processing is made at an early processing stage. THE PROBLEM OF BREAKTHROUGH Many experimenters began to discover evidence that the filter was not preventing the activation of semantics. some participants in dichotic listening tasks did notice if their name was presented in the unattended message. This effect is called breakthrough of the unattended message. A different way of accounting for semantic processing of the unattended message was proposed by According to their theory, all inputs are analysed for meaning before selective attention operates. The bottleneck in processing is at the point of response selection, where only the most important signals switch in other processes such as memory storage or motor output. This is the classic late selection model When the prime was followed by a different type of mask, which did not contain letter fragments but was a pattern of random dots (a noise mask), there was no associative priming. Marcel argued that this result provided evidence for two different types of masking; one type produced by a noise mask which degrades the perceptual input at an ‘early’ stage of processing and prevents information being passed to an identification OBJECT SELECTION, INHIBITION AND NEGATIVE PRIMING Evidence for semantic processing of unattended stimuli is also found in experiments using negative priming. Negative priming refers to the finding that the response time to categorize a target item will be slowed if that same item has been presented on the previous trial as a distractor item which was to be ignored. Clearly then, the distractor item must have been semantically processed despite the instruction to ignore it. DIRECTING THE SPOTLIGHT OF VISUAL ATTENTION The spotlight analogy suggests that objects or events in the beam of visual attention will be highlighted for processing and easily detected, whereas those outside the beam will not be. When we search the visual environment we can make an eye movement, or saccade, to direct the focus of visual attention to a location if there is something there we wish to know about. This movement is overt, and it is obvious to others where you are looking. However, we may also attend to something ‘out of the corner of our eye’, without making an eye movement. If you fixate your gaze here – *** – you are still able to make judgements about the page layout, surrounding colours, shapes and so on. In this case you are intentionally directing, or orienting, your attention in an endogenous way, using top-down control. This type of orientation of attention is covert (hidden), because where your eyes are directed does not reveal where you are attending. However, sometimes, despite intentionally attending to something, for example reading in the garden, the attention spotlight may be captured exogenously by a sudden movement or change in the visual environment such as understanding the processes underlying the orienting of visual attention, and measured the effects of providing participants with a visual cue which indicated the probability of the location where a visual target would appear. Two types of cue were used; a central cue, presented at the central fixation point, which was an arrow pointing left or right; or a peripheral cue, which was a brief illumination of a box presented to the left or right, in peripheral vision away from the fixation point. The task was simple. Participants fixated the centre of the visual display and responded as quickly as possible when they detected a target which would be presented either to the left or right side of the fixation point. No eye movements were allowed, so attention would have to be covertly directed. Recent evidence has begun to suggest that central cueing by an arrow can sometimes trigger automatic attention shifts even if it is uninformative, and also that the orienting of visual attention can be triggered by the direction of gaze of a face presented at fixation. This phenomenon is known as gaze- mediated orienting and allows us to share attention with a person we see looking at something. Shared attention is one example of the role of orienting attention in everyday life. CROSS-MODAL CUEING OF ATTENTION So far we have been concerned with attention processes within a single modality, such as vision or hearing. However, we live in a world of objects that have both visual and auditory properties, as well as other properties such as what they feel, smell or taste like. Experiments on attention have investigated cross-modal effects on orienting spatial attention, for example between seeing and hearing. participants could split auditory and visual attention in certain conditions, but when targets were expected on the same side of space for both modalities, the orienting effects were greatest.  VISUAL SEARCH What about searching a cluttered visual environment? Often we are looking for something specific, such as a product on the supermarket shelf or a familiar face in a crowded street. And although we are surrounded by multiple sources of sensory information we observe a coherent scene – the traffic may be moving around us but we see ‘a red bus moving’, ‘green grass below’, and ‘blue sky above’ rather than ‘red sky moving’ a ‘blue bus above’ and ‘green sky below’. We shall now consider experiments addressed at understanding how attention is involved in visual search and that aim to explain how one object can be selected from amongst other distracting or irrelevant objects An important step toward understanding this was taken by who proposed a feature integration theory (FIT) of attention. According to FIT, attention is the ‘glue’ that sticks the features of objects together. We have already mentioned the binding problem, which is concerned with how different properties of a stimulus are correctly combined in the introduction and in studies of subliminal priming. The initial assumption of FIT was that different sensory features are coded by specialised independent sub- systems or ‘modules’ and each module forms a feature map for the dimensions of the feature it encodes. Colours are represented on the feature map for colour, while lines of different orientations are represented on the orientation map. Detection of individual features that are represented on all the different maps takes place without the need for attention, in parallel, but if a conjunction (or binding) of features is required the separate features must be combined. A conjunction of features can be achieved three ways. First, features that have been encoded on the feature maps may fit into predicted object ‘frames’, according to stored knowledge; for example we know that buses are red and the sky is blue, so that if the colours red and blue are active at the same time, EVIDENCE FOR AND AGAINST FIT conjunction search the time taken to find a target increased linearly with the number of distractors in the display. However, if the target was defined by a unique feature, search time was independent of the number of distractors. This difference in search times was taken as evidence that when a conjunction is required to identify the target, a serial search is necessary in which focal attention moves to each object location in turn, conjoining features at each location until the target is found, when the search stops. This is called a serial self-terminating search. However, when a target is defined by just one distinctive feature, that feature is available on its feature map and calls attention to itself, resulting in pop-out. THE IMPORTANCE OF TASK DIFFERENCES difference between experiments supporting early selection and those suggesting late selection is due to differences in the overall demand for attention in an experimental task. They distinguished between selective filtering and selective set paradigms. Tasks such as dichotic listening or conjunction search involve selective filtering, which is more complex and attentionally demanding than a selective set task such as As selective filtering tasks tend to provide evidence for early selection and selective set tasks tend to provide evidence for late selection, perhaps the nature of the task is an important factor in how attention seems to operate You should find that shifting goals between tasks, updating and monitoring are very demanding on attention and short-term working memory. the behaviour of patients with frontal damage, who demonstrate frontal lobe syndrome COMBINING TASKS In everyday life we are familiar with the problem of trying to do two tasks at once and know that sometimes we have to stop doing one task because it is impossible to do them both without making a mistake. Often we find ourselves having to ‘multitask’