What is the mechanism for the Dows process

Involuntary attention

The selection of information is a process of paying attention to events or stimuli that provide information.

One distinguishes between

  • arbitrary attention
  • (top-down process): there is a selection intention, e.g. B. when looking for an object; and
  • involuntary attention
  • (bottom-up process), which is caused not by a previous intention of a person, but by events in the environment, e.g. B. a loud bang

The involuntarily triggered attention is only related to the process of selecting information, i. H. it is externally controlled (tied to exogenous events), but also implies internal (endogenous) factors so that selection becomes possible.

Two different selection processes can trigger involuntary attention:

  • The specific selection is based on a specific, latent pre-setting of the person, which leads to certain contents being selected and involuntarily attracting attention (e.g. an appetizing dish when someone is hungry). In contrast to arbitrary attention, where there is a willed (explicit) selection intention.
  • The unspecific selection refers to the deviation from other features characteristic of the situation in question; there are (involuntary) stimuli to choose from, the specific characteristics given by the context Not (e.g. a vertical bar under loud horizontal bars = pop-out effect, a term from TREISMAN, 1982)
  • Explanations for the registration of the situational deviations that trigger attention processes are provided by so-called Level jumps and Breaking the rules.

    A Level jump is a sudden change in a stimulus situation, e.g. B. a new stimulus appears or a characteristic of a stimulus changes and takes on a new value (especially investigated in acoustics). It is assumed that this is a simple sensory mechanism for registering deviations from a current situation, which reacts to discrete changes in sensory characteristics.

    A Breaking the rules occurs when the regularity of a previously existing sequence of events is broken, e.g. B. a dripping faucet that stops dripping. It is believed that a more complex mechanism capable of registering deviations from a pre-existing event structure is in place for this type of situational change.

    The two mechanisms differ in their coding (sensory dimension vs. sequential structure) and in their duration (a level jump is registered more quickly than a rule break).




    Level jumps and involuntary attention


    Orientation reaction


    : An orientation reaction (OR) is a syndrome of physiological and behavioral reactions that occur when there are changes in the stimulus constellation.

    These reactions include e.g. B. the alpha blockade in the EEG, the drop in skin resistance, as well as eye and head movements in the direction of the place where the change takes place. The OR is an indicator that changes in the situation (e.g. level jumps) have been registered by the organism.

    SOKOLOV (1963)

    explains the origins, habituations (habituation) and dishabituations ("weaning") of the OR with his Concept of the "neural model of the stimulus". Properties of a permanent or repetitive stimulus are stored in the form of a neural representation of the stimulus (e.g. intensity, color, stimulus duration, etc.); if a current stimulus situation no longer matches the neural representation, an OR is carried out.

    Event-related potentials

    Event-related potentials are brain electrical potential changes that occur in connection with certain events (e.g. environmental stimuli). The ERP research (ERP = event-related potential) provides findings, according to which level jumps lead to characteristic changes in the ERP and are related to involuntary attention. An ERP shown in the voltage-time diagram consists of several negative and positive maxima (for clarification: negative deflections are on the positive X-axis, positive deflections are on the negative X-axis). These components are described by specifying their polarity (negative or positive) and the number of deflections in this direction or the latency of the maximum deflection. After a level jump, an N 1 (first negative deflection) and then a P 2 (second positive deflection) are often observed. This appearance suggests that the N 1 generator processes taking place in the auditory cortex are responsible for triggering attention (NÄÄTÄNEN, 1990). NÄÄTÄNEN formulated the hypothesis that on z. T. subcortical level, certain stimulus properties are extracted, which, however, are not yet consciously perceived. A level jump activates systems that send interrupt signals to central processing mechanisms in which N 1 generator processes that trigger attention are involved.

    Response time studies

    Attention-giving processes cannot always be driven by external behavior, such as B. eye or head movements can be observed, as they are mostly limited to internal processes. POSNER (1980) describes these processes as "covert orientations of attention".

    Numerous reaction time studies have been carried out to investigate the covert attentional phenomena. POSNER, NISSEN & OGDEN (1978) dealt with the question of whether people can change the direction of their attention independently of the respective direction of fixation.

    Experiment by POSNER, NISSEN & OGDEN:

    The subjects are given a fixation point in the visual field. An arrow pointing to the left or right is displayed in this area. This serves as a reference stimulus for the target stimulus presented after a variable time interval, to which the person should react by pressing a button. The control group received no cues. The result was a significantly shorter reaction time of the VG to stimuli that were in the direction previously indicated by the arrow, as well as a significantly higher reaction time in the VG when the stimulus was presented in the opposite direction of the arrowhead. The cue stimulus (arrow) triggered a concealed turn of attention. The subjects turned their attention in the direction indicated by the arrow even before the target stimulus was presented.

    This investigation involved an arbitrary application of attention, as it was triggered by the intention of the subject.

    Experiments by YANTIS & JONIDES:

    To determine the situational triggers of involuntary processes, JONIDES (1981) used a similar experiment. The subjects had to find a target stimulus from letters arranged in a circle. Approx. 100 - 50 msec before the circular letters were presented, a cue was offered that was located in the middle of the display (central) or outside the circle (peripheral). The result was surprising: the peripheral cue could not be suppressed willingly, even when the instruction was given to ignore the cue. JONIDES & YANTIS drew the conclusion from their experiments that peripheral cues become triggers for involuntary attention-giving processes when they occur abruptly.

    MÜLLER & RABBITT (1989) examined the time frame in which involuntary attention-giving processes take place. They postulate based on their results two independent attention-giving mechanisms:

    • a faster mechanism that is triggered by a peripheral cue and is responsible for involuntary attention, and
    • a slower mechanism triggered by a central cue that is responsible for voluntary attention.

    More recent studies by YANTIS & JONIDES (1990) and THEEUWES (1991) contradict these two independent mechanisms. An attention-giving process triggered by a sudden change in the situation is not controlled completely externally, but is also dependent on internal factors. Their results show that an abrupt stimulus does not trigger an involuntary attention-giving process if the attention is already directed by a previous cue.

    It remains to be clarified which connections exist between the situational variables (time interval between cues and target stimuli, validity of central and peripheral cues) and the degree of attention focus, and whether the difference between voluntary and involuntary attention is not due to a single process mechanism that is based on differences in the triggering conditions.

    Breaches of rules and involuntary attention

    Orientation reaction

    An OR takes place not only on level jumps, but also on rule breaks. A rule break occurs when a sequence of already presented stimulus combinations is changed. The stimulus combination can consist of different stimulus modalities, such as B. a sound-light pair. A study by BADIA & DEFRAN (1970) showed that breaking the rule leads to an OR. They offered a sound stimulus 15 times followed by a light stimulus. The 16th presentation, which only contained either a sound stimulus or a light stimulus, triggered an OR (cf. classical conditioning by PAWLOW).

    These findings imply a mechanism that is able to detect deviations in the context of regular events; H. to distinguish stimuli that fit in the sequence from stimuli that do not fit.

    Event-related potentials

    In experiments to investigate attention processes using event-correlated potentials, the "Oddball Paradigm" used. This is the presentation of stimulus sequences consisting of standard stimuli with a high probability of occurrence and other stimuli (so-called deviants) with a low probability of occurrence. Involuntarily triggered attention-giving processes are examined using the passive oddball paradigm, whereby the subject's attention is diverted from the stimuli that are actually of interest. This happens because the subject is supposed to work on a task that typically involves a different sensory modality (e.g. reading and ignoring auditory stimuli when auditory stimuli are examined).

    NÄÄTÄNEN (1988) postulates another mechanism that can trigger involuntary attention in the case of auditory stimuli. The MMN (mismatch negativy), part of the N 2 component, is visible as a negation of the divergent stimulus compared to the standard stimulus and, in the case of discrete stimuli, can distinguish significantly better between stimulus repetition and stimulus change. Compared to the standard stimulus, the different stimulus can show changes in intensity, in duration of stimulus, in the location of the sound transmitter, in the interstimulus interval or in frequency. The following hypothesis provides a possible explanation for the occurrence of MMN: the physical parameters of a standard stimulus are decoded into a neural representation, with which each incoming stimulus is compared. If the result is a discrepancy, an MMN occurs, which in turn triggers an OR. The situations that trigger an MMN in the passive oddball paradigm sometimes also require a positive response after about 300 msec (P 3).

    Experiment by SQUIRES et al. (1977)

    In this study, in which combined light-sound stimuli were presented (e.g. blue light with 1000 Hz sound, orange with 1100 Hz, blue with 1100 Hz and orange with 1000 Hz), bimodal interferences were found. In one condition, subjects were required to count visual deviants and ignore acoustic stimuli, and in another condition the reverse was true. As expected, the deviants in the directed modality triggered a P 3, but also deviants in the disregarded modality. On the basis of these results, it seems that attention is paid by non-action-relevant deviants. The P 3 component could be understood as an indicator of a distraction of attention from the primary task.

    The effect of physical and semantic deviations on ERP components was examined several times.

    BESSON & MACAR (1987

    ) were able to observe the effect of deviating from more complex regularities in their experiment. Stimulus series of 7 stimuli each were presented in 4 different conditions:
    1. 6- or 7-word sentences that were visually presented word by word
    2. geometric patterns of increasing or decreasing size in the course of the series of stimuli
    3. Tones on a scale with increasing or decreasing frequency over the course of the series of stimuli
    4. Phrases from familiar melodies

    Test design: 25% of the stimulus series ended with an incongruent (non-matching) stimulus, whereby in

    • Condition 1 a semantically incongruent word appeared,
    • Condition 2 a pattern that violated the implicit rule of changing the size of the stimulus series,
    • Condition 3 a wrong frequency in the stimulus series and
    • Condition 4 a tone not belonging to the melody served as incongruent stimuli.

    In conditions 2, 3 and 4, incongruent stimuli were followed after 350-450 msec

    P 3 complex. In condition 1, the P 3 component was superimposed by a negation of the N 400. The authors suspected that the first words of each sentence could lead to a semantic priming of the words, and that pre-activated words could be processed more easily. If the primed semantic category deviates, this is reflected in an N 400 component.

    The visual search (see previous presentations) is a continuous selection process in which rules are broken. It is only worth mentioning that the visual search involves unspecific selection processes based on internal models. However, these mechanisms are still largely unknown.


    The term involuntary attention is intended to make it clear that it is not a person's intention that leads to attention, but that it is triggered by stimuli from the outside world.

    A specific selection leads to an arbitrary application of attention. A selection occurs unspecifically if environmental events are defined by their relationship to the situational context, and it triggers involuntary attention-giving processes.

    Viewed in this way, an involuntarily triggered attention-giving process differs from an arbitrary, intentional one only in relation to its cause. The question of whether these are two independent mechanisms or whether a common mechanism is activated in different ways by existing stimulus classes remains open.