How do we process information? What depends on whether we select one stimulus and not others? How can we attend to only one stimulus among many?
From basic psychology, many authors have focused on attention processes, proposing theoretical models of it. Today we are going to meet one of them, the Broadbent rigid filter model.
To do this, let’s start by seeing what the attention filter models are, to which this one by Donlad Broadbent belongs.
Filter models of attention focus on the concept of filtering. This consists of choose a fragment of the sensory stream and give it access to the central processing channel while the rest are lost (selective attention).
There are two types of filter models, which are as follows.
Precategorical filter models
Broadbent rigid filter model belongs to this type. In these models the selection of information is early, that is, the attentional mechanism operates in the initial phases of processing.
The process consists of a sensory registration (which is carried out in parallel, and where the physical characteristics of the stimulus are analyzed), the appearance of the filter (which selects the information it serves) and finally the appearance of a single central channel where the stimulus arrives. to be processed semantically (serially).
In these models, only properly attended information is processed at a semantic level.
Here the selection of information is late, that is, the attentional mechanism operates later than in the previous one.
In these models, there is a sensory store that includes an analyzer system (processing of physical and semantic characteristics in parallel and automatically/passively). Subsequently, the filter appears, which collects all the signals and selects them.
The next element is short term memory or central attention mechanism, which actively and consciously analyzes the message (it is a controlled process, which consumes attentional resources).
In these models, unlike the previous ones, all information is processed at a physical and semantic level.
The Broadbent Rigid Filter Model
This is a precategorical filter model, where the filter comes before semantic analysis. That is, the stimuli appear first, which will be stored in the sensory warehouse. Then the filter would act, which would select the information.
This would be stored in short-term memory (a channel of limited capacity), and finally certain information would pass to long-term memory (between these two elements, the subject’s responses to the stimulation would appear).
Something similar to the “law of all or nothing” would apply here, that is, the information passes or it does not pass (one message is processed at a time).
Broadbent Rigid Model Features
The processing would be done in parallel at the peripheral level.
Sensory memory temporarily retains information. The filter proposed is rigid and selective, since it chooses a fragment of the sensory flow and gives it access to the central channel, while the rest of the non-relevant information is lost (this occurs to avoid overloading of the central channel caused by the channels). multiple sensory).
At a central level, a categorical or semantic analysis of the information is carried out, that is, sequential processing occurs and the channel has limited capacity.
Broadbent Filter Features
It is a rigid filter, an “all or nothing” device. The selection of information is made based on physical characteristics of the stimuli without taking into account the meaning of the messages. Semantic processing will be done later.
On the other hand, the probability that a message is selected or not will depend on the properties of the stimuli (spatial location, intensity, speed of presentation, sensory modality…) and the state of the organism.
The filter can only focus on one channel or message at a time and its transition is two seconds.
The Deutsch and Deutsch early selection model
It is important to distinguish between Broadbent’s rigid filter model and Deutsch and Deutsch’s early selection model. The latter, unlike Broadbent’s rigid model (precategorical), is a postcategorical filter model.
In this case, we ask ourselves the following: how is it possible to select one input among several without analyzing them all? So that it can be decided which of the stimuli is relevant some degree of analysis is necessary.
Furthermore, in this case there is semantic analysis data prior to selection, which is handled by the analyzer.
Once the stimuli are received, are stored in the sensory warehouse. Subsequently, an analyzer acts prior to the filter, and the latter will transfer the information to the short-term memory (STM).
Finally, from MCP it is transferred to long-term memory (and, as in Broadbent’s model, responses will appear between these two memories, in the transfer from one to the other).
That is, the most important difference with respect to Broadbent’s rigid filter model is that in that of Deutsch and Deutsche there is an analyzer that acts prior to the filter.
