Reason, J. (1984). Lapses of attention

Reason, J. (1984). Lapses of attention Group 1: Ruth Buchan Based on a statement by William James (1890) that proposes "habit diminishes the conscious attention with which are acts are performed," this chapter explores the interaction between control of highly routinized activities and the amount of attention necessary to ensure they are successfully completed. Slips and lapses occur everyday and are not bizarre or random events, rather they follow an obvious pattern and do not depend on circumstance or the person performing the action. James and Jarrow suggest actions that are habitual are carried out by cognitive mechanisms with only a minimum of conscious involvement, and so unless there is some kind of intentional involvement the action will follow a familiar pattern. Slips of action; circumstances and frequency of involvement. One method of studying lapses of atention is through asking participants to keep a diary of their slips of action. Although this method is open to a number of biases, it is useful as the person carrying out the action is able to provide the most information about it. Reason (1979) carried out an extended diary study with 63 undergraduates, asking them to note and answer questions about every slip they made over a seven day period. the data showed that slips commonly occur when carrying out highly familiar tasks requiring little cose attention in highly familiar circumstances. The slips were shown to rapidly increase during the early hours of the waking day, decline betweeen noon and 3pm and peak between 5pm and 7pm. This shows errors are associated with periods of maximum activity. Error proneness questionaires given to the participants showed perceived liability to commit a particular type of error could be used as a predictor of their tendency to commit other errors. Classifying slips of action. Analysis of the slips of action committed by the participants suggests most errors can be assigned to one of four behavioural categories; repetition, wrong object, intrusion or omission. categorising errors reduces them to a amanagable level and agreement between judges as to which category an error belongs to has shown to be high. Theoretical considerations. To go beyond the behavioural level of analysis it is necessary to understand the mechanisms that govern human action. A model of human action is presented with intention, action and need systems linked with verbal and nonverbal memory, and affected by input and output functions and internal / external feedback. According to this model, errors in the same behavioural category have different underlying mechanisms. Four broad and overlapping categories of cognitive failure can be identified; control mode failures, intention system (and associated store) failures, action system failures and input function failures. Varieties of cognitive failure. Within each of the four types of cognitive failure listed above there are a series of hypothesis relating to specific error types. Two kinds of control mode failure are double capture slips and place losing / place mistaking errors. the first type occur because the intention system is captured at a crucial moment during the action and so the action system continues along a habitual course of action. The second occurs because atention is drawn to a largely automatic activity at an inappropriate time. Two kinds of intention systems failure are detatched intentions and lostintentions. In the first case an intention becomes misapplied to something other than intended, possibly because the intention system is only partialy involved in the action and so not processed at a sufficient enough level to last until the action has been successfully executed. lost intentions occur when the original intention is completely forgotten because of a complete breakdown at one stage during the execution of the action. Action systems failures include blends and behavioural spoonerisms, external activation of action schema and program counter failures. these errors occur because two action systems are operating at once, contextual cues cause a direct connection between input function and action system that bypasses the intention system or because the action system fails to keep track of the point reached in the instructional sequence. Finally, two kinds of input function failures are perceptual confusions and local unawareness. the first occurs because another object is used instead of the intened object because it looks the same as, is in the same place as or does a similar job as the intended object. The second occurs because preoccupation, abstraction or distraction means full conscious attention is not given to one's bodily state. Group 2: Kate Hill This chapter looks at the interaction between the control of highly routinized activities and the deployment of attention necessary for these activities to be successfully completed. Fundamentally, this is the relationship between attention and skilled performance. It focusses on the categorisation and potential explanations of everyday slips and lapses - when the intended action, for whatever reason, is not performed. The Extended-Diary Study An 'elaborate' diary study was carried out using 63 undergraduates who had to try to record all instances where their behaviour had deviated from their intention. It was hypothesised that 'the performance of highly automatised tasks in relatively predictable and familiar surroundings liberates the central processor from moment-to-moment control' - leading to focal attention either being captured by some other (parallel) mental activity or some other external event. This could then lead to the central processor failing to switch back to the original task at some 'critical decision point'. The results collected show quite a restricted variety of error-producing activities and their associated mental and physical conditions. Slips were shown to occur while carrying out highly familiar tasks that require little attention. The main categories were identified as i) strong-habit intrusions - where the resulting actions clearly belong to some other activity, ii) omissions, ii) repetitions, and iii) use of the wrong objects. The slips were made during periods of maximum activity in the day and there is some hint that they are most likely to occur in the transition periods between home and work. An error-proneness questionnaire was given to 85 undergraduates to assess the frequency of slips and lapses in everyday life. Slips were divided into two sections - those where the action deviated from intention (action slips) and those where 'lapses of memory' were more to blame (non-action slips). Non-action slips were clearly reported to occur more frequently than action slips. Classifying Slips of Action Slips can be defined by the behaviour which is outwardly seen or by assumptions about the cognitive processes which have brought them about. Four behavioural categories are suggested from the examination of 625 slips: 1) Repetition: some actions in the intended sequence are repeated unnecessarily. 2) Wrong object(s): intended actions made to wrong object(s). 3) Intrusion: Unintended action become involved in sequence. 4) Omission: Intended actions are left out of sequence. There are overlaps in these areas however - for example, all slips could be described as 'omissions' as they all involve a particular action failing to take place at the planned time. The evidence also suggests that these categories do not map on to corresponding cognitive failures - members of the same behavioural classes of slips can be mediated by different cognitive failures. A Heuristic Model of Action A model is given (p.533) suggesting some of the underlying mechanisms to human action. Importance is given to the 'intention system' and its links to the accessible (to consciousness) memory system. Other links to the intention system involve the internal state (needs) and external events (the input function). Possible Categories of Cognitive Failure 1) Control mode failures: errors made due to being in the wrong control mode. Mistakes are made, for example, when the intention system is involved in the ongoing activity of a task which requires it to undertake high levels of decision making. Alternatively, the intention system may be closed off from the ongoing activity while carrying out a highly automated task - another undesirable state. These failures are broken down into two sections: i) Double-capture slips, ii) Place-losing and place-mistaking errors. 2) Intention System failures: Those slips associated with the formation of plans and with the storage and retrieval‎ of plan-related information. These are closely related to the memory system and are broken down into two sections: i) Detached intentions, ii) Lost intentions. 3) Action System Failures: Mistakes caused by the unintended triggering of action schemata, there are three categories: i) Blends and behavioural spoonerisms, ii) External activation of action schemata, iii) Program counter failures. 4) Input Function Failures: Mistakes which are attributable to the input function - for failures in stimulus recognition or being aware of one's surroundings. These are divided into two sections: i) Perceptual confusions, ii) Local Unawareness. Group 3: Stuart Payne The aim of this chapter is to explore the interaction between the contrast of highly routine activities and the development of attention necessary to ensure their successful outcome. Reason uses various examples to back up his claim that lapses or slips are "not bizarre or random events", he claims that they "follow a closely discernable pattern that is largely independent of their perpetrator or the surrounding circumstances". He includes examples (such as the one from the spectator ) of people doing strange things and not noticing themselves. Circumstances and Frequency of Occurance Reason points to Oppenheim (1966) and his caution about the Volunteer bias, Selection bias, Recording bias or diary studies before describing Reason 1979. The study found that errors are more likely in familiar tasks which require little attention. Peak times (5-7pm) were found for erros and an error-proneness quesitonnaire showed that if people perceive themselves likely to commit one kind of slip, then they are in fact more likely to commit another kind. Classifying Slips Reason found from his diary study that 4 groups could be devised: (1) Repetition (2) Wrong Object (3) Intrusion (unintended action) (4) Omission (intended action not done) Theoretical Considerations Reason presents a slight adaptation of his 1977 model of the mechanisms governing human action - the heuristic model of action. The model contains: (1) Need System (a) Input Function (2) Intention System (b) Output Function (3) Action System (4) Memory System Together with INTERNAL and EXTERNAL feedbacks. Varieties of Cognitive Failure The failings characterised by slips should, according to reason, bring insight into cognitive functions. Here Reason attempts to show the range of failures that occur: CONTROL MODE FAILURE - Double Capture Slips - Place-losing and Place-mistaking slips INTENTION SYSTEM FAILURE - Detached Intentions - Lost Intentions ACTION SYSTEM FAILURE - Blends and Behavioural Spoonerisms - External activation of Action Schemata - Program Counter failures INPUT FUNCTIONS FAILURES - Perceptual Confusions - Local Unawareness Group 4: Sunita Rattan HABIT AND ATTENTION Both habit and attention have leading parts to play in the guidance of action. In a novel situation you have to be more attentive with increasing practice it shifts to one of habit EVERDAY SLIPS AND LAPSES/ SLIPS OF ACTION James and Jastrow indicate slips frequently occur as a result of misdirected or diminished attention. Dirary keeping technique method used to keep a record of how these actions occur. However caution should be taken. 3TYPES OF BIAS TO REPORING ACTION SLIPS 1)Volunteer bias 2)Selection bias 3)Recording bias EXTENDED DIARY STUDY (REASON 1979) hypotheses to be tested 1) Notion that the performane of a highly automatized task in relatively predictable and familiar surroundings liberates the central processor from moment to moment control. As a result- focal attention tends to be "captured" by some pressing but parallel mental activity or by some unrelated external event. DISTRIBUTION OF SLIPS OVER HOURS OF THE DAY Suggests errors are associated with periods of maximum activity Also hint that slips are most likely to occur during transistions between hm and wk. (when people either depart for wk or return hm) These distractions appear to reflect influnce of situational factors rather than what is known about durnal variations in efficiency Findings also show slips of action occur during exection of highly familiar tasks regarding little in way of close attention RELATIVE FREQUENCY ERROR TYPES Perceived liability to commit certain kinds of say action slips is a better predictor of one's tendency to perform other kinds of deviant actions than it is of one's proness to nonaction lapses and vice versa Slips of action can be readily classified into 4 behavioural categories 1)Repetitions 2)Wrong objects 3)Intrusions 4)Omissions on the basis of error statements provided by the subjects Examinations of error corpus suggests that explanations of slips and lapses could be organized around 4 theoretical mechanisms 1)Control mode failures 2)Intention system failures 3) Action system failures 4)Input function failures Neither evidence nor the theoretical assumptions are sufficient to permit the unique determination of the causes of any one error A particular behaviuor error form could arise from a variety of possible cognitive failures and different behavioural categories of action slips could be due to the same type of failure. APPLIED PSYCOLOGISTS Task is to prevent error but when some catastrophic lapse does occur HUMAN FACTORS SPECIALIST Asks what knowledge of cognitive processes can reveal about the orgins of this particular disaster. -------------------------------------------------------------------------------- Reason, J. (1984). Lapses of attention Group 1: Ruth Buchan Based on a statement by William James (1890) that proposes "habit diminishes the conscious attention with which are acts are performed," this chapter explores the interaction between control of highly routinized activities and the amount of attention necessary to ensure they are successfully completed. Slips and lapses occur everyday and are not bizarre or random events, rather they follow an obvious pattern and do not depend on circumstance or the person performing the action. James and Jarrow suggest actions that are habitual are carried out by cognitive mechanisms with only a minimum of conscious involvement, and so unless there is some kind of intentional involvement the action will follow a familiar pattern. Slips of action; circumstances and frequency of involvement. One method of studying lapses of atention is through asking participants to keep a diary of their slips of action. Although this method is open to a number of biases, it is useful as the person carrying out the action is able to provide the most information about it. Reason (1979) carried out an extended diary study with 63 undergraduates, asking them to note and answer questions about every slip they made over a seven day period. the data showed that slips commonly occur when carrying out highly familiar tasks requiring little cose attention in highly familiar circumstances. The slips were shown to rapidly increase during the early hours of the waking day, decline betweeen noon and 3pm and peak between 5pm and 7pm. This shows errors are associated with periods of maximum activity. Error proneness questionaires given to the participants showed perceived liability to commit a particular type of error could be used as a predictor of their tendency to commit other errors. Classifying slips of action. Analysis of the slips of action committed by the participants suggests most errors can be assigned to one of four behavioural categories; repetition, wrong object, intrusion or omission. categorising errors reduces them to a amanagable level and agreement between judges as to which category an error belongs to has shown to be high. Theoretical considerations. To go beyond the behavioural level of analysis it is necessary to understand the mechanisms that govern human action. A model of human action is presented with intention, action and need systems linked with verbal and nonverbal memory, and affected by input and output functions and internal / external feedback. According to this model, errors in the same behavioural category have different underlying mechanisms. Four broad and overlapping categories of cognitive failure can be identified; control mode failures, intention system (and associated store) failures, action system failures and input function failures. Varieties of cognitive failure. Within each of the four types of cognitive failure listed above there are a series of hypothesis relating to specific error types. Two kinds of control mode failure are double capture slips and place losing / place mistaking errors. the first type occur because the intention system is captured at a crucial moment during the action and so the action system continues along a habitual course of action. The second occurs because atention is drawn to a largely automatic activity at an inappropriate time. Two kinds of intention systems failure are detatched intentions and lostintentions. In the first case an intention becomes misapplied to something other than intended, possibly because the intention system is only partialy involved in the action and so not processed at a sufficient enough level to last until the action has been successfully executed. lost intentions occur when the original intention is completely forgotten because of a complete breakdown at one stage during the execution of the action. Action systems failures include blends and behavioural spoonerisms, external activation of action schema and program counter failures. these errors occur because two action systems are operating at once, contextual cues cause a direct connection between input function and action system that bypasses the intention system or because the action system fails to keep track of the point reached in the instructional sequence. Finally, two kinds of input function failures are perceptual confusions and local unawareness. the first occurs because another object is used instead of the intened object because it looks the same as, is in the same place as or does a similar job as the intended object. The second occurs because preoccupation, abstraction or distraction means full conscious attention is not given to one's bodily state. Group 2: Kate Hill This chapter looks at the interaction between the control of highly routinized activities and the deployment of attention necessary for these activities to be successfully completed. Fundamentally, this is the relationship between attention and skilled performance. It focusses on the categorisation and potential explanations of everyday slips and lapses - when the intended action, for whatever reason, is not performed. The Extended-Diary Study An 'elaborate' diary study was carried out using 63 undergraduates who had to try to record all instances where their behaviour had deviated from their intention. It was hypothesised that 'the performance of highly automatised tasks in relatively predictable and familiar surroundings liberates the central processor from moment-to-moment control' - leading to focal attention either being captured by some other (parallel) mental activity or some other external event. This could then lead to the central processor failing to switch back to the original task at some 'critical decision point'. The results collected show quite a restricted variety of error-producing activities and their associated mental and physical conditions. Slips were shown to occur while carrying out highly familiar tasks that require little attention. The main categories were identified as i) strong-habit intrusions - where the resulting actions clearly belong to some other activity, ii) omissions, ii) repetitions, and iii) use of the wrong objects. The slips were made during periods of maximum activity in the day and there is some hint that they are most likely to occur in the transition periods between home and work. An error-proneness questionnaire was given to 85 undergraduates to assess the frequency of slips and lapses in everyday life. Slips were divided into two sections - those where the action deviated from intention (action slips) and those where 'lapses of memory' were more to blame (non-action slips). Non-action slips were clearly reported to occur more frequently than action slips. Classifying Slips of Action Slips can be defined by the behaviour which is outwardly seen or by assumptions about the cognitive processes which have brought them about. Four behavioural categories are suggested from the examination of 625 slips: 1) Repetition: some actions in the intended sequence are repeated unnecessarily. 2) Wrong object(s): intended actions made to wrong object(s). 3) Intrusion: Unintended action become involved in sequence. 4) Omission: Intended actions are left out of sequence. There are overlaps in these areas however - for example, all slips could be described as 'omissions' as they all involve a particular action failing to take place at the planned time. The evidence also suggests that these categories do not map on to corresponding cognitive failures - members of the same behavioural classes of slips can be mediated by different cognitive failures. A Heuristic Model of Action A model is given (p.533) suggesting some of the underlying mechanisms to human action. Importance is given to the 'intention system' and its links to the accessible (to consciousness) memory system. Other links to the intention system involve the internal state (needs) and external events (the input function). Possible Categories of Cognitive Failure 1) Control mode failures: errors made due to being in the wrong control mode. Mistakes are made, for example, when the intention system is involved in the ongoing activity of a task which requires it to undertake high levels of decision making. Alternatively, the intention system may be closed off from the ongoing activity while carrying out a highly automated task - another undesirable state. These failures are broken down into two sections: i) Double-capture slips, ii) Place-losing and place-mistaking errors. 2) Intention System failures: Those slips associated with the formation of plans and with the storage and retrieval‎ of plan-related information. These are closely related to the memory system and are broken down into two sections: i) Detached intentions, ii) Lost intentions. 3) Action System Failures: Mistakes caused by the unintended triggering of action schemata, there are three categories: i) Blends and behavioural spoonerisms, ii) External activation of action schemata, iii) Program counter failures. 4) Input Function Failures: Mistakes which are attributable to the input function - for failures in stimulus recognition or being aware of one's surroundings. These are divided into two sections: i) Perceptual confusions, ii) Local Unawareness. Group 3: Stuart Payne The aim of this chapter is to explore the interaction between the contrast of highly routine activities and the development of attention necessary to ensure their successful outcome. Reason uses various examples to back up his claim that lapses or slips are "not bizarre or random events", he claims that they "follow a closely discernable pattern that is largely independent of their perpetrator or the surrounding circumstances". He includes examples (such as the one from the spectator ) of people doing strange things and not noticing themselves. Circumstances and Frequency of Occurance Reason points to Oppenheim (1966) and his caution about the Volunteer bias, Selection bias, Recording bias or diary studies before describing Reason 1979. The study found that errors are more likely in familiar tasks which require little attention. Peak times (5-7pm) were found for erros and an error-proneness quesitonnaire showed that if people perceive themselves likely to commit one kind of slip, then they are in fact more likely to commit another kind. Classifying Slips Reason found from his diary study that 4 groups could be devised: (1) Repetition (2) Wrong Object (3) Intrusion (unintended action) (4) Omission (intended action not done) Theoretical Considerations Reason presents a slight adaptation of his 1977 model of the mechanisms governing human action - the heuristic model of action. The model contains: (1) Need System (a) Input Function (2) Intention System (b) Output Function (3) Action System (4) Memory System Together with INTERNAL and EXTERNAL feedbacks. Varieties of Cognitive Failure The failings characterised by slips should, according to reason, bring insight into cognitive functions. Here Reason attempts to show the range of failures that occur: CONTROL MODE FAILURE - Double Capture Slips - Place-losing and Place-mistaking slips INTENTION SYSTEM FAILURE - Detached Intentions - Lost Intentions ACTION SYSTEM FAILURE - Blends and Behavioural Spoonerisms - External activation of Action Schemata - Program Counter failures INPUT FUNCTIONS FAILURES - Perceptual Confusions - Local Unawareness Group 4: Sunita Rattan HABIT AND ATTENTION Both habit and attention have leading parts to play in the guidance of action. In a novel situation you have to be more attentive with increasing practice it shifts to one of habit EVERDAY SLIPS AND LAPSES/ SLIPS OF ACTION James and Jastrow indicate slips frequently occur as a result of misdirected or diminished attention. Dirary keeping technique method used to keep a record of how these actions occur. However caution should be taken. 3TYPES OF BIAS TO REPORING ACTION SLIPS 1)Volunteer bias 2)Selection bias 3)Recording bias EXTENDED DIARY STUDY (REASON 1979) hypotheses to be tested 1) Notion that the performane of a highly automatized task in relatively predictable and familiar surroundings liberates the central processor from moment to moment control. As a result- focal attention tends to be "captured" by some pressing but parallel mental activity or by some unrelated external event. DISTRIBUTION OF SLIPS OVER HOURS OF THE DAY Suggests errors are associated with periods of maximum activity Also hint that slips are most likely to occur during transistions between hm and wk. (when people either depart for wk or return hm) These distractions appear to reflect influnce of situational factors rather than what is known about durnal variations in efficiency Findings also show slips of action occur during exection of highly familiar tasks regarding little in way of close attention RELATIVE FREQUENCY ERROR TYPES Perceived liability to commit certain kinds of say action slips is a better predictor of one's tendency to perform other kinds of deviant actions than it is of one's proness to nonaction lapses and vice versa Slips of action can be readily classified into 4 behavioural categories 1)Repetitions 2)Wrong objects 3)Intrusions 4)Omissions on the basis of error statements provided by the subjects Examinations of error corpus suggests that explanations of slips and lapses could be organized around 4 theoretical mechanisms 1)Control mode failures 2)Intention system failures 3) Action system failures 4)Input function failures Neither evidence nor the theoretical assumptions are sufficient to permit the unique determination of the causes of any one error A particular behaviuor error form could arise from a variety of possible cognitive failures and different behavioural categories of action slips could be due to the same type of failure. APPLIED PSYCOLOGISTS Task is to prevent error but when some catastrophic lapse does occur HUMAN FACTORS SPECIALIST Asks what knowledge of cognitive processes can reveal about the orgins of this particular disaster. -------------------------------------------------------------------------------- 링크: <a href="http://www-users.york.ac.uk/~nvh1/Skills/summaries99.html#T2.2" target=nlink>http://www-users.york.ac.uk/~nvh1/Skills/summaries99.html#T2.2</a>