- 1 Definitions of Attention
- 2 Controlled vs. Automated Processes
- 3 Selective Vs. Divided Attention
- 4 Other Theories of Attention
- 5 Bibliography
Definitions of Attention
- Means by which we actively process a limited amount of information from the myriad of information available through our senses, stored memories, and other cognitive processes.
- Refers to the allocation of processing resources assuming limited resources
- The ability to focus on a given task->Focusing on certain features of the environment to the exclusion of other features
- The ability to concentrate->Allows either selective awareness or responsiveness
Controlled vs. Automated Processes
Controlled processes are those which a person actively partakes in, the processes that require a person to focus or concentrate are considered controlled processes. In terms of attention as the allocation of processing resources assuming limited resources, controlled processes consume a very large amount of resources. Automated processes are those which require very little focus, concentration, or attentional resources, depending on the definition used. These processes can interfere, support, or have no effect on one another. When learning a new process, an old automated process can prevent learning-learning to swing a tennis racket after learning to swing a baseball bat- advance learning- throwing almost any object- or do nothing- knowing proper running form will not affect the previous two tasks.
All processes begin as controlled processes. Most tasks that are performed are a series of mental processes. When a process is considered a controlled process, it is performed serially, every step of the process must be performed one at a time until the controlled process finally becomes an automated process through automatization. Controlled processes are basically newly learned processes.
This is the name for the process through which a process changes from a controlled to an automated process. As a controlled process is repeated several times, the neurons which fire in the process eventually move into position so that the current that flows through them meets very little resitance, thus streamlining the process. As the neurons rearrange themselves, the process becomes automated.
These are the processes which can be performed in parallel, multiple different processes at one time. These processes require very little attention and occur automatically, sometimes without the knowledge of the person performing the task.
Selective Vs. Divided Attention
Selective attention is when a person attends to one task over others.
Divided attenion is when a person attends to multiple tasks simultaneously. This kind of attention has been shown to be affected by three factors: task difficulty (performance and task difficulty have an inverse relationship), task similarity (performance and task similarity also have an inverse relationship), and practice ( the more practice, the better the performance).
Dichotic Listening Task
- Auditory Theory of Attention(Cherry 1953; Moray 1959)
- Subjects were instructed to listen to 2 different conversations and 'shadow' one of them. This forced the subjects to focus on only one of the conversations. In the end, the subjects were able to recall information about the conversation that was listened to with no out of the ordinary results, but the participants could only tell physical charateristics of the speaker of the unattended message. The subjects could only state the gender, whether the sound was human or a noise, and the distance of the speaker. Information about the actual message could not be recalled including the language spoken and the actual words spoken, even if the message was nothing more than the repition of one word. The experiments did show that people will change the focus of their attention when a certain stimulus is used. The effect has come to be called the cocktail party phenomenon because of how when in a conversation, with noise in the background, a person will respond to the call of their name that has was said in a tone that no one else heard.
Broadbent's Filter Theory
This filter theory is based on Cherry's studies and states that as information is processed, there is a filter which blocks the processing of most stimuli in an effort to keep the the processing mechanism of the brain from overloading. According to Broadbent, this filter blocks all but the physical characteristics of a stimulus from being processed if another stimulus is being processed at that time. This theory, however, does not account for a person's ability to increase their ability to attend to multiple stimuli.
Evidence against Broadbent’s Filter Theory
Moray (1959) showed that about one third of the time a subject could detect their own name in the non-shadowed message. This anomaly shows that there is some semantic filter that Broadbent never accounted for.
Underwood (1974) showed that subjects who had previously participated in shadowing tasks could eventually be trained in dichotic listening tasks and recall about 67% of digits from the non-shadowed message, whereas non-experienced subjects could only recall about 8%.
Triesman's Attenuation Theory
Triesman basically revised Broadbent's theory. Triesman's attenuation theory states that people process stimuli in an hierarchy from attended to unattended stimuli and attentional resources are allocated accordin to this hierarchy. As the processing becomes deeper, the information processed bottlenecks.
Evidence for Treisman’s Attenuation Theory
Gray and Wedderburn (1960) played the messages “What 6 hell” and “2 the 9” and the subjects reported hearing “What the hell” and “2 6 9”. Their study shows that there is a semantic filter that Broadbent’s theory does not account for in his theory. Also, shows that it is easier for subjects to switch between messages than Broadbent concluded.
Visual Search is the finding of a target amidst the environment. Visual search is affected by the similarity of distracters and the target and the set size being searched.
Theories of Visual Search
Ulric Neisser believed that subjects automatically process the basic features of a scene in parallel. He showed that subjects more easily located a target object when it was around dissimilar objects rather than similar. For example, it is easier to find the letter “X” when it is surrounded by the letter “O” than when it is surrounded by the letter “K”.
Other Theories of Attention
Feature Integration Theory
Triesman developed the theory of Feature Integration which states that all stimuli are first processed in parallel and then serially in a conjuntion search in which certain features are looked for in combination. This theory has led to other theories such as guided search theory, but ignores the effects of the similarities between distracters and the target. The theory also incorrectly assumes that processing occurs in two phases: parallel, then serially.
Guided Search Theory
Developed by Wolfe in 1998, Guided Search theory is a revised version of Feature Integration theory. Guided Search theory states that serial and parallel processing occurs simultaneously in differing amounts. At first an 'activation map' is created and objects that are similar to the target are identified. Next, the similar objects are processed serially while all other objects are processed in parallel to dientify the target.
Central Capacity Theories
Central capacity theories state that the ability for someone to attend to a stimulus is based on the amount of attentional resources that the person has. Problems with this theory mainly involve how the resources are distributed: theoretically, if one pair of tasks is similar but different and another pair is completely different, performance should be the same for both sets of tasks which is not how the experiments turn out.
Filter theories assume that attending to multiple tasks simultaneously results in only retaining information about certain characteristics in the varying tasks. Filter theorists believe that there is just a wall which blocks the further processing of information of all but one or two attended to tasks. This type of theory does not account for the ability of someone to improve their ability to attend to multiple tasks.
Bottleneck theories state that information is processed simultaneously but the processing of information decreases in performance as the number and difficulty of the tasks increases.
"A Level Psychology Through Diagrams." Grabame Hill. Oxford University Press: New York City, 2001.
All information came from Mr. Fischer's class notes.