Please enable JavaScript.
Coggle requires JavaScript to display documents.
Alternative Models: Cognitive Energetic Model (Cognitive Energetic Model…
Alternative Models: Cognitive Energetic Model
Problems with the Arousal Theory
What is the difficult task? Hard to define the requirements
methodology: additional effect of distracting attention from the task used to measure performance efficiency
can accommodate almost any pattern of results --> impossible to falsify
concept of arousal: there should be different types!
inability to define effects of stressors on arousal level independently of effects on performance
Psychophysiological objections
Measures: EEG, muscle tension & galvanic skin response (somatic level), heart rate HR, blood pressure & pupil diameter (autonomous nervous system level), catecholamine & cortisol excretion (endocrine level)
Paradoxical rapid eye movement sleep REM → one is asleep but measures such as HR, muscle activity & EEG show high level of arousal
lack of correlation among psychophysiological indexes during performance eg. HR deceleration during the foreperiod in a choice reaction. HR decelerated and muscle tension increased while preparing to respond. → indecisive
physiological variables have other functions unrelated to arousal --> always reflect a mixture of specific, perceptual-motor, and energetic effects
unusual psychophysiological patterns eg. increase rather than a decrease of HR and muscle activity as a function of time on task in a monotonous setting
Behavioral Objections
counteracting effects of arousing and dearousing variables? Broadbent's criticism
noise only had detrimental effects during the later part of a work , but arousal theory predicts a stronger effect at the beginning
the effect of heat, which is supposed to be arousing as well, did not interact with the effects of the other energetic factors in the same way as noise
the inverted-U pattern of interactions was only found in repetitive and monotonous tasks such as vigilance and self-paced serial reactions
Sternberg: Additive Factor Model
determining which stages are involved in a particular information processing task
If two factors have additive effects (i.e., if the effect of one factor does not depend on the level of another factor), they are assumed to affect different stages
If the factors interact, such that the effect of one factor depends on the level of the other, the two factors are assumed to affect the same stage
stage robustness: if a number of experiments produce data consistent with the proposed stages
uses factorial designs & RT data
Levels of a factor: the factor "gender" has the levels M (male) and F (female)
there is a condition for each possible combination of factor levels
The effect of a factor F that has two levels, F1 and F2, in a reaction-time experiment is the difference between the RTs at its two levels: RT(F2) - RT(F1)
The method is limited to discovering modules that are arranged in series, as a sequence of stages BUT some mental processes could occur in parallel!
Why should we expect complex processes to have parts? --> "principle of modular design" by Marr
otherwise: a small change in one place will have consequences in many other places, the process as a whole becomes extremely difficult to debug or to improve
improves our understanding of the information processing
Criteria for separation of parts
Detachability: A part can be detached from the rest relatively easily; the subparts of a part are more coherent than is one part with another.
Replaceability: A part can be replaced by a new part (usually in the same class) without causing large changes in the rest of the object.
Locality: different parts occupy non-overlapping regions of space
Independence (separate modifiability): One part can be modified without altering the other parts
Functional distinctness: Different parts do different things, have different purposes
Cognitive Energetic Model (Sanders)
different stages of information processing have different relations to effortful processing
different basal energetic mechanisms/systems: arousal, activation, effort
Arousal similar to phasic arousal (rapid, transient reactions to transient situations)
Activation similar to tonic arousal (varies more slowly, internal state of the organism)
sleep loss seems to affect both arousal and activation
Effort: consistent with Broadbent's upper system, minimizes damage of adverse conditions
supervision: supplying additional energy to the arousal and activation systems when they fall short and dampening these systems when they tend to flood
only invoked to force performance up (not a normal condition)
Sanders varied stimulus quality and S-R compatibility in a CRT task.
conditions carried out following a night’s loss of sleep or after normal sleep
interaction of the effects of sleep loss and stimulus quality. Sleep loss had a more pronounced negative effect on CRT when stimuli were degraded than when they were intact
variation of S-R compatibility proved insensitive to sleep loss
the effect of sleep loss was not determined by complexity per se but by the nature of the cognitive demands
Qualitative changes in energetic state
Stresses such as noise and sleep loss would not affect the level of arousal or the amount of capacity but would lead to a qualitatively different distribution of energetic resources needed to comply with the task demands. → resource strategy approach -> central control that aims at performance stability by comparing actual performance with internal reference values
general aim of any human being is to restore an energetic equilibrium when disturbed by a stress --> human organism is a strategic, self-regulating system reacting to deviations
Performance properties
selectivity: related to relative priority of tasks in dual-task performance, eg. Central stimulus tracking suffers from sleep loss vs. Peripheral stimulus tracking suffers from noise
speed & accuracy: Shifts in SAT toward speed in conditions of high and toward accuracy in conditions of low arousal. (Sleep loss led to more missed stimuli and longer CRTs, whereas noise had the effect of a higher error rate)
general alertness: reflected by performance variability--> performance is less variable in an alert state
short-term memory capacity: after sleep loss, participants retrieved earlier presented items at the cost of more recent ones. In contrast, under noise, they retrieved the most recent ones at the cost of earlier presented digits
AFM stage structure but an extension: computational stages are supposed to receive, and cannot function without, energetic supply from two basal energetic systems
Abnormal conditions affect energetic supply because either too little or too much is supplied
too little supply of activation is supposed to be harmful to the level of preparation and to readiness for action
too much supply (computational systems flooded) when facing emotional or threatening stimuli. Eg. Performance under stress
evaluation system that collects information about whether the flow of ongoing satisfactory (criteria vary as a function of motivational state)
response selection is not affected by the basal mechanisms but depends directly on effort (most speculative part of the model)
gradually varying degree of energetic supply
Eg. conditions of too high level of arousal/activation --> disorganization of performance when effort fails to dampen the oversupply
Information processing approach
unobservable processes by which information in a stimulus is translated into a response
traditional measures: speed & accuracy
new measures: psychophysiology & neuroimaging
Basic model: 3 processing stages
Perception/ Stimulus identification
decision making and response selection (or stimulus-response translation)
response programming and execution
results of one stage of processing form the input to the next one
Information Theory: human is not just a receiver of information, but also a transmitter of information → human as info channel → we can talk about the rate of information transfer and the efficiency of transmitting information → quantify the information, after which we can examine the time it takes to receive and transmit a given amount of information
if no reduction of uncertainty has taken place, no information has been transmitted
The amount of information in a stimulus depends in part on the number of possible stimuli that could occur in a given context
if different stimuli are possible, there will be some uncertainty about what will happen next, and the stimulus will convey information
amount of information in a stimulus expressed in bits
Human processing efficiency often described as the rate of information transmission expressed in bits per second
Hick-Hyman law: reaction time (RT) in a task is linearly related to the amount of information transmitted
suppose that information is processed in a series of discrete stages or modules
Mental Chronometry: measurement of the time needed to complete various processes
Donders’ subtractive method: assuming that processing stages are independent, assumption of pure insertion → compute the time needed to complete a known process
Behavioral Measures
Reaction Time RT
from the onset of stimulus presentation until the participant’s response
Influencing factor: readiness of the responder
Accuracy
proportion/percentage of correct responses
The more alternatives, the lower the percentage correct
speed-accuracy trade off function (S-Shaped): RT and accuracy are inversely related: increases in response speed are generally accompanied by reduced accuracy of performance
Asking subjects in an experiment to “respond as quickly as possible without making too many errors” is asking them to find an optimal point on the speed-accuracy trade-off function
Unidimensional arousal
Yerkes & Dodson: inverted U curve
Easterbrook: relations between complexity & arousal
As the level of arousal increases, intake of both task-relevant & task-irrelevant cues narrows
As the level of arousal decreases, relevant and irrelevant cues are readily admitted (widening).
at a low level of arousal, the main problem is distractability and competing response tendencies due to irrelevant cues
at a high level of arousal, insufficient task-relevant informa¬tion makes it impossible for reliable action
a more complex task, characterized by more task-relevant features, should suffer more from relatively high arousal than a less complex task
Objections: Eysenck
narrowing may be strategical rather than determined by deficient cue utilization
High arousal is sometimes thought to be related to high anxiety → it’s not that wider cue utilization means more openness to experience
Experiment: RT analysis of the effect of levodopa (using the additive-factor method) --> which information-processing stages are mediated by dopamine?
two-choice visual RT task
3 variables: Signal intensity, stimulus-response mapping and foreperiod duration --> additive effects on RT--> 3 independent stages
Visual signal intensity affects stimulus preprocessing stage
stimulus-response compatibility affects response selection stage
duration of the foreperiod affects the motor adjustment stage
Results
shorter mean RTs under levodopa (236 ms) than under placebo (246 ms)
strong signal intensity (227 ms) yielded shorter mean RT than the weak one (255 ms)
compatible stimulus-response mapping yielded shorter mean RT (216 ms) than the incompatible one (266 ms)
faster mean RT for the short (224 ms) than for the long (258 ms) foreperiod
One significant interaction: effect of signal intensity on mean RT was larger under placebo (31 ms) than under levodopa (25 ms)
single 200-mg dose of levodopa speeds up choice RT in healthy subjects without increasing the frequency of errors