Attention

Chapter 2: Attention Peter Naish  1 Auditory attention 37 auditory 34, 37–45  1.1 Disentangling sounds 37  disentangling sounds 37–41 Important theorists - Important theories - Important effects   shadowing (head shadowing) 39–40the ear further from the sound is slightly shadowed by the head, so receives a some what quieter sound.  The head is not a very large obstacle, so the intensity difference between the ears is not great.   The waves pass by almost as if the head was not there.  This means that there is negligible intensity shadowing, so the intensity cue is not available for direction judgement with low-pitched sounds.  wavelengths 39–41 The distance from one wave crest to the next (low pitch = long wavelength).   1.2 Attending to sounds 41   Broadbent, D.E. 41, 42, 43, 44, 50, 58 spotlight of attention 41 The way in which we can bring our attention to bear on a desired part of the environment (almost like moving eyes). dichotic listening 41, 43, 44, 54, 60, 61 Presenting listeners with as simplified version of the real world of multiple sounds.  Participants wear headphones and receive a different sound in each ear often different recorded, spoken messages. echoic memory 41 From dichotic listening - the store’s quality of hanging onto a sound for a short time like a dying echo. serial information capture 42,43             like a computer all information is processed one piece at a time parallel information capture 42 Broadbent’s (1954) theory was that, after the first early stage of parallel information capture, a ‘gate’ was opened to one stream of information and closed to the rest. aircraft pilots and attention 42, 46, 64 pilots had trouble picking out a radio message from a number of interfering stations. The headphones delivered the same signals to each ear, so it was not possible to use inter-aural differences to direct attention to the wanted message. Broadbent devised a stereo system, which played the desired signal through both headphones, while the interference went only to one or the other. This made the interference seem to come from the sides, while the signal sounded as if it was in the middle (identical waves at the two ears).   1.3 Eavesdropping on the unattended message 43 cocktail party effect 43 If you are attending a noisy party you can be attending to one conversation but then someone mentions your name – meaning is extracted from unattended material.  If you have previously selected your conversation to focus upon and serially process how can other sounds be processed enough to register? Treisman, A. 43–4, 49, 50, 55                attenuation process 43 Functioning as if there was a filter which turns the volume down for all but the attended signal. Thus while most goes unnoticed we are particularly sensitive to some signals like our name so it would still be processed and attract our attention. shadowing (dichotic listening) 43,61 dichotic listening technique which requires the participant to repeat aloud everything that is heard in one ear, following like a shadow close behind the spoken message. (an example of priming). Priming 43 Sensitization of a temporary kind as in Treisman (1960). Lexical decision task 43 Supports priming. The task asks parts to say as quickly as poss. whether a string of letters spell a real word.  People respond quicker if the word is preceded by another related to it e.g. doctor after nurse compared to after cook. Corteen, R.S. and Wood, B. (1972) 44                GSR (galvanic skin response) 44, 144, 467,500 The sweat glands of the nervous person begin to secrete and the salty fluid lowers the resistance to small non-shocking current.  This change is the galvanic skin response (GSR) and is used in lie detectors. Deutsch, D. and Deutsch, J.A. 44 suggested that all messages received the same processing, (in parallel) whether they were attended or not Norman, D.A. 44, 325, 336, 357, 572 proposed that unattended information must at least receive sufficient processing to activate relevant semantic memories   2 Visual attention 45   visual attention 45–54 and hearing 45–6 and seeing 45–6                there is a similarity between hearing and seeing aircraft pilots and attention 42, 46, 64 Military jets are often flown very fast and close to the ground (to avoid radar detection), requiring the pilot to attend intently to the outside view.  To avoid the pilot having to look down into the cockpit, the ‘head-up display’ (HUD) was developed.   2.1 Knowing about unseen information 46   Sperling, G. 46–7, 49 partial report superiority effect 47 In the test the parts could recall the same number of items from a cued row of letters as they could from the whole set of 12 when uncued.  Suggests that more is processed than suggested by the whole set test. iconic memory 47 Equivalent to echoic memory in hearing. backward masking 47, 49, 50 (challenge to sperling) The visual equivalent of shadowing (to prevent conscious processing of material) in vision. SOA (stimulus onset asynchrony) 47,50 The time between onset of the target display and the onset of the mask Evett, L.J. and Humphreys, G.W. (1981) 47, 48, 49, 64 used stimulus sequences containing two words, both of which were masked. The first was supposed to be impossible to see, while the second was very difficult. It was found that when the second word was related to the first (e.g. ‘tiger’ following ‘lion’) it was more likely to be reported accurately; the first, ‘invisible’ word apparently acted as a prime. Cheesman and Merikle, (1984) P.M. 47–8, 54 pointed out that although participants say they cannot see masked words, they often do better than chance when forced to guess whether or not one had actually been presented. Priming and unseen information 48–9 more recently researchers have provided persuasive evidence that meaning can be extracted from material of which the participant is unaware. Pecher, D. 48 Pecher et al. (2002) used the Evett and Humphreys (1981) technique, but with two  modifications. First, the priming word could be displayed either for a very short time, so that it was allegedly undetectable, or it was shown for a duration of 1 second, giving ample time for reading and guaranteeing a priming effect.  In the 1 sec./10% related condition, there was a 15% priming advantage, which rose to 40% for 1 sec./ 90% related. This jump can be explained by participants guessing the target, often correctly. With very brief presentation of the prime, the priming advantage of 90% over 10% related was smaller. Guessing was not possible here, so the small priming advantage suggests the primes must have been sufficiently analys to activate their meaning.   2.2 Towards a theory of parallel processing 49   backward masking 47, 49, 50 Coltheart, M. (1980) 49, 54, 208, 209, 629, 641–2,645 Coltheart (1980) offered an elegant solution to the problem of why participants did not know information, such as whereabouts in the display the letters occurred, built around the semantic/episodic distinction. Coltheart proposed that items do not normally reach conscious awareness unless both the semantic and episodic detail are detected. These different kinds of information have to be united, and this process, Coltheart (1980) suggests, takes time and attention. According to this account, Sperling’s 12 letters, or even Evett and Humphrey’s lion, are indeed processed in parallel to cause semantic activation, but the viewer will not become aware of this, unless able to assign the corresponding episodic details.   2.3 Rapid serial visual presentation 50 Turvey (1973) 50 masking Integration 50 – when the SOA between target and mask is very short so that the two items are perceived as one. Interruption 50 – occurs at longer SOAs and the interruption masking is experienced even if mask and target are presented to different eyes. Koelers (1968) central (interruption) masking 50 Masking occurring at the level of central decision mechanisms. Introduced in terms of the concurrent and contingent model of Turvey (1973). peripheral (integration) masking 50 According to Turvey (1973), masking occurring at the level of peripheral encoding mechanisms. backward masking 47, 49, 50 Broadbent, D.E. and Broadbent, M.H.P. (1987) 41, 42, 43, 44, 50, 58 Display of a series of stimuli in rapid succession so that each served as a backward mask for the preceding item. Rapid Serial Visual Presentation (RSVP) 50–4 In RSVP, a series of up to 20 stimuli is presented in rapid progression so each serves as a backward mask for the item before.  Identification is hard; participants are usually asked to identify only two.  Each item is a single black letter, except for the one white letter.  After seeing the sequence, the participant must identify the white letter (t!), and say whether there was an X in the list (T2), Typically, T2 is more likely to be identified if it is immediately after T1, but optherwise identification is unlikely unless 5 or 6 items separate T1 and T2. Vogel, E.K. 51, 52 Vogel et al. (1998) conducted an RSVP experiment that used words, rather than single letters. Additionally, before a sequence of stimuli was presented, a clear ‘context’ word was displayed, for a comfortable 1 second. For example, the context word might be shoe, then the item at T2 could be foot. However, on some presentations T2 was not in context; for example, rope. attentional blink 52 awareness of T2 droops for several items after T1 as the sysrem needs  time (attentional blink – the AB effect) before it can process something new. However, the AB effect vanishes wwhen the item after T1 is omitted, and when T2 is the last in the listm so it can be seen if it is not masked. refractory period, 52 Seems to be this gap between processing and processing something fresh. (In vision the refractory period is referred to as this or AB.) Raymond, J.E. 52 Raymond et al. (1992) used a typical sequence of RSVP stimuli, but omitted the item immediately following the first target. In other words, there was a 100 ms gap, rather than another item following. Effectively, this meant that the degree of backward masking was reduced, and not surprisingly resulted in some improvement in the report rate for T1. Very surprisingly, it produced a considerable improvement in the reporting of T2; the AB effect had vanished Giesbrecht, B. and Di Lollo, V. ()52, 106, 107 Gisebrech and DI Lallo explain the AB effect, through their two-stage model: parallel processing of target characteristics, follwed by serial processing of information, preparing it for awareness and report.   2.4 Masking and attention 54 Merikle and Joordens (1997) Addressed whether material available for processing, but happened not to be at the focus of attention and whetrher the fate of material that a participant was trying to attend to, but did not have time to process were related. they characterized it as a distinction between perception without awareness (such as in masking studies) and perception without attention (as with dichotic listening). They concluded that the results were entirely comparable, and that the same underlying processes are at work in both kinds of study.   3 Integrating information in clearly-seen displays 55 3.1 Serial and parallel search 55   Treisman, A. 43–4, 49, 50, 55 serial and parallel processing, in visual displays 55–6 Treisman pointed out that the first two displays have target items which differ from the rest on only one dimension; to identify the third target it is necessary to consider two dimensions. Treisman argued that all the different items are processed at the same time, in parallel. Treisman proposed that it is also required to link simple features. Each item in the display has to receive attention just long enough for its two features (shape and colour) to be combined, and this has to be done one item at a time until the target is found. in visual displays, the number of distractors does not affect the time taken to decide whether a target is present, but this takes longer for two dimensions.  Different types of analysis, e.g. shape and colour, take place in different parts of the brain, so must be processed serially. Attention is needed to integrate information from separate sources.   brain damage and visual displays 56 Treisman (1998) also reports investigations with a patient who had suffered strokes in that region. He was shown simple displays, containing just two letters and asked to describe the first letter he noticed in the display. He had correctly identified features that were present, but was unable to join them appropriately. The implication of this is that both the detection and the integration of features are necessary steps in normal perception, and that integration requires attention.   3.2 Non-target effects 56 Duncan, J. and Humphreys, J.W. (1989) 56, 57   Treisman 56–9 feature integration theory 56–9 and non-target effects 56–7 triesman’s feature integration theory is challenged by the T/L display of Duncan and Humphreys (1989). It took longer to find L in the display when Ts were in both horizontal and vertical orientations, suggesting that non-targets must be recognised and reject, as well as find the target.   3.3 The ‘flanker’ effect 57             Shaffer and LaBerge (1979) found priming effects, even when they presented words in a way which might have been expected to eliminate priming. ‘flanker’ effect 57–9 participants took longer to identify the category of a noun, e.g. animal or vegetable, when words from another category appeared above and below the target, suggesting these flankers were being processed in parallel, producing a slowing, similar to the stroop effect.    Broadbent and  Gathercole, S.E. 58, 59, 330 suggested that same-category words act as primes, consistent with feature integration theory. The Stroop effect Stroop (1935) reported a number of situations in which the processing of one source of information was interfered with by the presence of another. A variant is the ‘Emotional Stroop task’, which can be used in therapeutic diagnoses. Dudley et al. (2002) used colours to print a list of words, some of which were associated with negative emotions (e.g. the word sadness). Depressed people have an attentional bias towards such depression-related material. Patients were required to name the ink colours for each word, as quickly as possible. Both depressed patients and those in the early stages of Alzheimer’s disease were slower than a control group, but only the patients with depression were extra slow in responding to negative words. The technique permits an appropriate diagnosis.   flanker asymmetry effect After Hostmann et al. (2006), where responses to a positive expression flanked by negative expressions suffer from more interference than do responses to a negative expression flanked by positive expressions. flanker interference effect In the flanker task, where flankers that are inconsistent with the target produce less accurate and/or slower  responses than flankers that are consistent with the target.   4 Attention and distraction 59 attention and distraction 59–62 distraction 59–62 music, and distraction 59, 61 parallel processing and visual information 49, 52, 59   4.1 The effects of irrelevant speech 60 serial recall 60 someone talking interferes with serial recall, but meaningful material (e.g. numbers when digits are to be recalled) is no more distracting than other items.  Speech appears to be distracting because it is not constant (jones, 1990)   dichotic listening 41, 43, 44, 54, 60, 61 In the context of dichotic listening, it was shown that ignored auditory material may nevertheless be processed, and hence its meaning influences perception of attended material. However, meaning appears to have no special impact, when speech interferes with memory for visually presented material. Thus, hearing numbers spoken, while trying to remember digits, is no more damaging than listening to other irrelevant speech items (Buchner et al., 1996).   Jones, D.M. and Macken 60, 327–8 Rhyming words (cat,hat) were less disruptive than cat, dog. Jones suggests we must organise sound strings, i.e. recognise items and note their order. This is simpler with rhyming words, so there is less disruption.  Hearing the letter m/k/l as three separate, unchanging streams of sound was less disruptive than as a repeated sequence of 3 items. The impact of irrelevant speech shows that parallel processing is not always possible when simultaneous demands are made on the same process of sequencing items.   4.2 Attending across modalities 61   Driver, J. 61, 63 ventriloquism effect 61 This is encountered at the cinema where the speakers are at the side of the screen yet the voice seems to come from the actor in the middle; we bind visual and acoustic information. Participants found it hard to shadow one of two messages spoken by the same person, with sound coming from the same loudspeaker when the image of the speaker appeared on a tv monitor, just above the loud speaker, but easier with the monitor to the side;  the two messages were the perceived as separate.  This has implications for the use of hands-free headsets when driving, with an earpiece in just one ear. The callers voice appears to come from just one side, so may distract the driver’s attention from what is ahead   5 The neurology of attention 62 attention and neurology of 62–5   5.1 The effects of brain damage 62 sensory neglect 62–4,73 damage to one of the parital lobes leads to sensory neglect, where a person does not respond to stimuli, e.g. food, on the ‘wrong’ side, and ignores the ‘wrond side of an onject when drawing.  It appears to be half the object rather than half the field of view which is neglected. Driver, J. and Halligan, P.W. 63 Driver and Halligan showed participants pairs of pictures of objects like chess pieces, and asked whether they were the same or different.  When the pictures were tilted, so the relevant feature was on the neglected side of the object, but in the non-neglected field of view, the patients did not respond to the difference. Bisiach, E. and Luzzatti, C. 63 A patient of Bisiach and Luzzatti demonstrated neglect when asked to describe what could be seen looking towards the cathedral square of the city where he grew up. When asked to describe what could be seen when looking towards the cathedral, he again only refered only to the right-hand side, reporting information which he had previously ignored, and not reporting information he had previously seen. Balint’s syndrome 64 Results from damage to both lobes. Patients have difficulty in shifting attention, e.g. one patient could not report the letters making up a word, since there was difficulty in switching from one letter to the next, but could read the whole word, when it was seen as a single item. (reminiscent of pilots and HUD)   Baylis, G.C. 64 Baylis et al. (1994) described a patient who could not report the letters making up an isolated word. Viewed in this way, each letter was a small object and it was not possible to switch attention from one to the next. dorsal stream, ventral stream and attention 64 Early visual processing takes place in two major pathways in the brain, known as the ventral and dorsal streams (Section 6 of Chapter 3); the parietal region is part of the dorsal pathway. Damage to the ventral stream results in different kinds of integration problems; patients are aware of all aspects of a scene, but to the patient they remain segmented into small elements. For example, an individual shown a photograph of a paint-brush described seeing a wooden stick and a black object (the bristles) which he could not recognize. Humphreys, G.W. 47, 48, 49, 64 Humphreys (2001) suggests that the varieties of different problems are evidence that the binding together of different features takes place in several different stages and brain locations.   5.2 Event-related potentials 64 ERPs (event-related potentials) 64–5 and studies of semantic processing 217, Electrical firing of neurons that can be recorded following receipt of a stimulus Woldorff, M.G. 64 examined ERPs evoked by sounds. These included signals occurring as soon as 10 ms after the auditory event.  The earliest stages of registration at the auditory cortex were detected after about 20–50 ms. It was of particular interest that, whereas the 10 ms signal was not affected by attention, the magnitude of the electrical activity in the cortex was smaller when the sounds were played to an unattended ear.  

New Page