DE102013017820A1 - Method and system for visualizing the emotional impact of visual stimulation - Google Patents

Abstract

The invention relates to a method and a system for visualizing the emotional effect of visual stimulation (1). The method comprises the steps of: stimulating a subject (2) with the visual stimulation (1), measuring the gaze behavior of the test person (2) during the stimulation, measuring at least one indicator of the emotional effect on the subject (2) synchronously with the subject Stimulation and detection of the gaze behavior, - determination of fixation data from the measurement of gaze behavior, which indicate which areas (11) of the stimulation by the subject (2) for which time intervals and / or how often have been fixed (fixation areas), - associate to the thus determined fixation regions (11) of data relating to the at least one indicator of the emotional effect, and - representation of the fixation regions (11) and their associated data concerning the at least one indicator of the emotional effect in a visualization map (10).

Description

The invention relates to a method and system for visualizing the emotional impact of visual stimulation.

In many areas, eg. As in advertising, e-commerce and in terms of the design of products, it is important to make means of communication, such as a billboard, a shop page on the Internet or products so that the behavior of viewers and potential consumers expedient being affected. Because only then is it likely that an advertisement to buy or a shop page makes visitors to buyers.

The emotional impact is an important predictor of behavioral effectiveness. Are relevant information taken into account within the environmental stimulation and can leave traces in the brain (attention)? Is the information perceived pleasant and in line with the needs (attractiveness)? Does the information about behavior (activation) stir?

The brain continuously analyzes whether individual elements of the environment are unimportant or useful. From this calculation, it generates emotional responses that indicate the emotional impact (and thus the behavioral impact) of the stimulation.

If it is possible to reliably and validly measure the emotional impact of a stimulation by capturing emotional indicators, this will allow behavioral prediction with respect to stimulation.

For example, if an ad is emotionally negative, the behavioral prediction may be made that the viewer will be less likely to purchase the advertised product. By changing the design of the advertisement, the probability of a desired target behavior can be increased. Accordingly, there is a need to capture an emotional impact of a visual advertisement.

The same applies to websites. A large part of the websites was designed with the aim of favoring a particular behavior of the visitors. This goal behavior (eg, buying a product) is influenced by the emotional impact of the page. If the site has a positive effect, it is more likely that the visitor will stay there longer and perform the desired action. Negative reactions lead to leaving the site, and in the worst case, the lost visitor becomes a competitor's customer.

The measurement of emotional impact allows the empirically based selection and optimization of the design of products and behavior influencing tools.

From the US 2008/0965468 A1 Methods for measuring the emotional response to visual stimuli are described. In doing so, certain areas of interest (AOI) are defined in advance and physiological parameters are measured in the defined areas of interest. The disadvantage of such a procedure is that an isolated analysis of a stimulus or a stimulation, namely in predefined areas of interest takes place.

It is an object of the present invention to provide a method and system for visualizing the emotional impact of visual stimulation by analyzing and visualizing the complete stimulus without spatial confinement. Furthermore, a corresponding visualization map of the emotional impact of a visual stimulation should be provided.

This object is achieved by a method having the features of claim 1, a system having the features of claim 15, a visualization card having the features of claim 16 and a computer program having the features of claim 17. Embodiments of the invention are specified in the subclaims.

Thereafter, the solution according to the invention provides a method for visualizing the emotional impact of visual stimulation comprising the following steps. A subject is stimulated with stimulation to be examined for a defined time. For example, the subject views an advertisement, web page, or video sequence for a defined time. During the stimulation, the gaze behavior (especially gaze movement and fixations) of the subject is measured. Furthermore, during the stimulation, at least one indicator of the emotional effect (eg a physiological, a behavioral or a psychological parameter) is measured on the subject.

The method according to the invention determines fixation data from the measurement of the gaze behavior. These indicate which areas of the stimulation the test subject has fixed as frequently and for which time intervals. Such regions are referred to below as fixation regions. Measuring the gaze behavior of a subject is also referred to as "eye tracking". By measuring the gaze behavior, it is possible continuously, the gaze position of the subject and thus the field of Stimulation that the subject is currently looking at. In the limiting case, such a fixation area is a dot or a pixel in the case of a display of the stimulation on a computer screen. However, the respective area may also be larger depending on the precision with which the gaze position is detected when measuring the gaze behavior. From the measurement of the gaze behavior, fixation data are thus determined which define certain fixation regions of the stimulation which the subject observes or fixes during the stimulation and which define the time intervals and frequencies with which the subject has fixed the respective areas.

The method according to the invention further provides that the fixation areas determined in this way are associated with data concerning the at least one indicator of the emotional effect (also referred to below as the "emotional indicator"). Thus, any fixation region of the stimulation detected during the measurement of the gaze behavior is additionally characterized by such data concerning the measurement of at least one emotional indicator.

The fixation areas and their respective associated data concerning the at least one emotional indicator are displayed in a visualization map, which is, for example, a two-dimensional representation on a visualization medium (computer screen, printout, etc.). The representation can be, but does not have to be superimposed on the actual stimulation. If the stimulation is, for example, the presentation of an advertisement on a computer screen, then the presentation of the fixation-oriented data can be superimposed on the emotional effect of this presentation or alternatively on a separate screen.

The said method is advantageously carried out on a plurality of test persons. In this case, the acquired measurement data are averaged in each case. There are several possibilities with regard to the type and timing of the averaging. For example, an averaging is done before the data is displayed in a visualization map. Alternatively, the visualization maps provided with the data of individual test persons are superposed on one another and thereby averaged.

The solution according to the invention enables a complete and practically limitless analysis and visualization of the emotional effectiveness of a visual stimulation. Under a visual stimulation in the context of the present invention, both a purely visual stimulation and an audio-visual stimulation understood.

According to the invention, it is neither necessary to determine in advance nor retrospectively certain areas of interest of the stimulation for the analysis. The fixation data and the data concerning at least one emotional indicator (eg physiological parameters) represent the emotional effect. They are spatially analyzed and visualized for the complete stimulus. In the case of displaying the visualization map on a computer screen, the indicator values are calculated for each pixel of the stimulus. The limit is in fact the pixel size of the stimulation. The visualization card therefore provides a complete, complete and limitless picture of the emotional impact of visual stimulation.

The solution according to the invention makes it possible to answer the following questions for a visual stimulation or scene (for example a website, a print advertisement or a TV commercial):
First, the question is answered as to which components or stimuli of a considered stimulation attract the attention of the observer or the subject. In other words, it answers what the viewer sees and what he does not see. This is determined by the fixation data indicating the fixation regions of the stimulation and the associated time intervals and viewing frequencies. The fixation data thus provides an indicator of the emotional impact of a visual stimulus that indicates the viewer's attention.

Second, the question of at least one other indicator of emotional effectiveness is answered. If you measure z. For example, facial muscle activity, it can be answered how attractive the fixation-related stimulation is. It is thus found an answer to what the viewer likes and what he does not like. The measurement of facial muscle activity, for example, by means of electrodes, with discharges from the corrugator Supercilii (frown) and from the zygomatic major (laughing muscle) are made. The muscle potentials are continuously recorded and digitized at a high sampling rate. Activity in the corrugator is associated with negative emotions and activity in the zygomatic is associated with pleasurable emotions. This relationship has been scientifically proven. In other words, the attractiveness of stimulation leads to involuntary changes in the basic tension of these muscles. The attractiveness of a stimulation is often referred to as valence (attitude, value, usefulness).

The facial muscle activity data thus provides a fixation-related indicator of the emotional impact of visual stimulation which indicates the attractiveness of the fixed areas (fixation areas) within the stimulation.

It should be noted that valence of stimulation (attractiveness) alternatively via a video-based facial muscle analysis (based on a "Facial Action Coding System", or automatically via webcam), via an EEG (eg, the frontal hemisphere asymmetry in alpha). Band), can be measured by a functional magnetic resonance tomography, by heart rate changes or by a pupil dilatation.

The solution according to the invention may also include other indicators of emotional effectiveness. About the measurement of z. B. Skin conductance changes can be answered, whether the fixation-related stimulation of a stimulation leads to a motivational activation. So the question is answered, what motivates the viewer to an action and what not. This is done via the data of at least one further emotional indicator (eg physiological parameter), which is, for example, a skin conductance, for example the skin conductance of the non-dominant hand of the subject. So it is scientifically proven that motivationally meaningful, d. H. "Emotional" stimuli, unlike unimportant stimuli, lead to a stronger basic activity of the autonomic nervous system, especially a stronger basic activity of the sympathetic nervous system. Its activation is accompanied by increased sweat gland activity on feet and hands: the stronger the activation potential of a current stimulation, the more the skin is moistened. The moisture penetration of the skin becomes indexable by the skin conductivity measurement. The current skin conductivity value is continuously recorded and digitized. Alternatively, motivational activation can be achieved by other physiological parameters such as pupil dilation, heart rate change, heart rate variability, a parameter measured by an electroencephalography device, a parameter measured by a functional magnetic resonance imaging device, and skin temperature of the subject (also video based or webcam ) are measured.

Another indicator of the emotional impact may thus reflect the fixation-related motivational activation of the viewer. The motivational activation of stimulation is often referred to as arousal (arousal).

The indicators of the emotional impact are visualized layer by layer in the visualization map by means of their respective representation form at a glance in their entirety. The basis of this representation forms in one variant of the stimulus (image, video) itself. The indicator layers are visualized on this basis. This makes it possible to grasp and understand the relatively complex relationships of the emotional impact at a glance.

The analysis and visualization according to the invention thus provides indicators of emotional effect which enable a behavioral prediction for the handling of the observer or the subject with a stimulation and the stimuli provided by the latter. If the measured emotional indicators are not conducive to the desired behavior, the stimulation (eg the design of the advertisement) can be changed and re-measured by means of the present method, if the recorded indicators of the emotional effect are now more favorable and one in the sense the provider of the stimulation have a more favorable behavior-influencing effect.

The solution according to the invention captures the emotional effect (with the indicators attention, and preferably with at least one of the indicators attractiveness and possibly activation) fixation-related and spatially for the entire stimulus, thereby allowing insights that neither from the isolated analysis of individual components or Stimuli of stimulation would still result from a holistic analysis of stimulation as the only stimulus.

To illustrate this advantage of the present invention, consider the example of a web page where the conversion rate of the buy button changes the conversion rate, i. H. the proportion of visitors who become customers should be increased. An isolated test of different buttons could, for example, show that an orange button has the desired emotional effect rather than a green button. However, this result does not mean that the orange button will have the same effect in the context of the web page with its other design features. Maybe the green button will do better here. On the other hand, if you only look at the global emotional impact of the entire webpage, you may not see any difference between the orange and green button variants. Only the analysis of the button within the website as a form allows a statement about which button color rather achieves the desired emotional effect.

Which stimuli of the stimulation are evaluated, decides not the examiner, but is determined by the behavior of the test person. This is important to the present invention. Accordingly, the solution according to the invention does not work in advance with the determination of certain areas of interest. Rather, according to the invention, the gaze behavior data are used to give the indicators of emotional effectiveness a spatial dimension (insofar as the data concerning, for example, the facial muscle activity and, for example, the data relating to another emotional indicator are assigned to specific fixation areas). The solution according to the invention thus provides insights about which areas of stimulation attract attention, and whether the fixed areas z. For example, they may be considered attractive and / or lead to motivational activation (arousal), taking into account the overall context and all spatial areas of stimulation.

According to one embodiment of the invention, the fixation data (indicator of attention) determines the visibility of the original stimulus. In the visualization of the data, this can be achieved in various ways, for example by a brightness-coded representation of the fixation data. Such a representation can in principle be self-referential, criterion-related or differential, as will be explained in more detail below. For example, a visibility manipulation is performed in that the basic stimulus is first covered with a colored (and / or hatched) semitransparent layer, and the transparency of this covering layer in fixed areas is increased depending on the frequency or duration of the fixation. As a result, the pixels of the underlying original stimulus appear more clearly at those locations which the subject has fixed (and / or fixed more often than the areas that a subject has only glanced at or not at all for a longer time interval.

According to a further embodiment of the invention, the data relating to at least one additional emotional indicator (eg attractiveness, eg indicated by measuring facial muscle activity and / or activation (arousal), eg indicated by skin conductance) are displayed in color-coded form. In this case, it may be provided, for example, that in the event that the data indicate an attractiveness of the fixations in the stimulation, a warm color such. B. magenta is used for color coding, while low attractiveness (or repulsion) in the other case by a rather cold color such. B. cyan is encoded.

According to a further embodiment of the invention, the data on the attractiveness (eg of the subject's facial muscle activity) and at least one further emotional indicator (eg activation by the test person's skin conductivity) are measured during the stimulation, wherein first data concerning the attractiveness and second data regarding the at least one further emotional indicator are measured. The first data is color-coded and the second data is represented by geometric shapes or contour lines.

Thus, for example, it may be provided that the data regarding the further indicator of emotional effect in the visualization map are represented by circles of different diameters, wherein, for example, the diameter indicates the degree of the motivational activation (arousal) provided by this indicator. The contour or thickness of the perimeter can also code for emotional activation. Additionally or alternatively, it can further be provided that the motivational activation is reproduced in the form of contour lines, wherein the contour lines indicate certain percentage ranges of such a maximum value starting from a maximum of the motivational activation.

According to a further embodiment of the invention, the different data (fixation data, data relating to the at least one emotional indicator) are superimposed in the visualization map. However, this is not necessarily the case. For example, it may alternatively be provided that the data of different emotional indicators together form a single color coding. It can also be provided that certain of the data records can be hidden on the visualization map in order to be able to recognize and record the other data records with greater clarity.

According to a further embodiment of the invention, the assignment of the data relating to the at least one emotional indicator to the fixation regions takes place with a time delay in the sense that data are assigned to a contemplated fixation region, which originate from a time interval at the time interval in which the considered fixation region the test subject has been fixed with his eyes, is time-shifted. In other words, in this embodiment of the invention, a time delay is taken into account between the perception detection (information acquisition) determined by measurement of gaze behavior and the onset of a change in the measured emotional indicators with respect to the stimulus. Thus, studies have shown that a stimulus not directly parameters such. B. affects the facial muscle activity or the skin conductivity, but this can be delayed. The time offset thus takes into account and corrects an inertia of the emotional response to a visually captured stimulus.

The time delay (delay) can basically be within a wide time interval between 0.01 s and 10 s. Investigations have shown that in the detection of Facial muscle activity, in particular a time shift between 0.4 s and 1 s, and in the detection of skin conductivity in particular a time shift between 1 s and 4 s with regard to said inertial offset of the emotional response to a visually detected stimulus can be beneficial.

A stimulation which is visualized with regard to its emotional effect with the method according to the invention can basically be a two-dimensional or a three-dimensional scene, for example a two-dimensional scene printed or reproduced on a screen of a computer, for example a web page or a print display. The individual regions of such a stimulation, which a subject observes for certain time intervals and which are determined from the gaze behavior measurement, are two-dimensional regions of the considered two-dimensional scene. Nevertheless, in the context of the present description, they are sometimes referred to as "spatial" areas. This is to be understood that a spatial area is a specific area within the two-dimensional scene.

Due to the independence of the method according to the invention from somewhat defined areas of interest, the method is equally applicable to a sequence of images (video sequence). In this case, the duration of a single image representation forms the lowest limit of the resolution.

The data streams of the fixation data and the data relating to the at least one emotional indicator are preferably recorded synchronized in time. The synchronization can be realized for example via marker signals. This temporal synchronization makes it possible to assign corresponding data of the at least one emotional indicator to the individual fixation areas and time intervals of the fixation data.

According to a further embodiment of the invention, it is provided that the fixation data are smoothed around a contemplated fixation area, in particular in such a way that a z. B. radially decreasing course of attention is assumed. This takes into account that during a fixation the brain does not exclusively take in information of a certain fixed point or pixel (foveal area), but has potentially access to the information around the fixed point or the fixed pixel (parafoveal area). Such smoothing or "smoothing" can also take place in the data relating to the at least one emotional indicator.

Based on the idea of foveal focusing, the surrounding pixels of a current fixation area are thus gradually attributed decreasing values. A corresponding smoothing can also take place with regard to the data relating to the emotional indicators (eg attractiveness and / or activation).

In accordance with a further embodiment of the invention, the data associated with a considered fixation area relating to the at least one emotional indicator are spatially and temporally integrated and displayed spatially and temporally integrated in the visualization map. In the visualization card thus z. B. in each fixation area spatially and temporally integrated value of z. B. attractiveness and / or activation. This improves the clarity of the presentation in the visualization map.

• – Mittel zum Bereitstellen einer visuellen Stimulation für eine Testperson,
• – Mittel zum Messen des Blickverhaltens der Testperson während der Stimulierung der Testperson,
• – Mittel zum Messen mindestens eines Indikators der emotionalen Wirkung an der Testperson während der Stimulierung der Testperson,
• – Mittel zum Bestimmen von Fixationsdaten aus der Messung des Blickverhaltens, die angeben, welche Bereiche der Stimulation die Testperson für welche Zeitintervalle und/oder wie häufig fixiert hat (Fixationsbereiche),
• – Mittel zum Zuordnen zu den auf diese Weise bestimmten Fixationsbereichen von Daten betreffend den mindestens einen Indikator der emotionalen Wirkung, und
• – Mittel zur Darstellung der Fixationsbereiche und der diesen jeweils zugeordneten Daten betreffend den mindestens einen Indikator der emotionalen Wirkung in einer Visualisierungskarte.

The system according to the invention for visualizing the emotional effect of a visual stimulation comprises:

• Means for providing visual stimulation to a subject,
• Means for measuring the gaze behavior of the subject during stimulation of the subject,
• Means for measuring at least one indicator of the emotional effect on the subject during the stimulation of the subject,
• Means for determining fixation data from the measurement of the gaze behavior, which indicate which areas of the stimulation the test person has been fixing for which time intervals and / or how frequently (fixation areas),
• Means for associating with the thus determined fixation regions of data relating to the at least one emotional impact indicator, and
• - Means for displaying the fixation areas and their respective associated data concerning the at least one indicator of the emotional effect in a visualization card.

The invention will be explained in more detail with reference to the figures of the drawing with reference to several embodiments. Show it:

1 a system for visualizing the emotional impact of visual stimulation;

2 an example of a visual stimulation in the form of an image;

3 the visual stimulation of the 2 with additional representation of a fixation region of the stimulation, which fixes a subject during a time interval of the stimulation;

4 the representation of 3 wherein the fixation areas are additionally assigned a visibility coding in the form of a brightness coding which represents an indicator for the attention of the observer (eg fixation duration and / or fixation frequency);

5 the representation of 4 in which, in addition, a color is associated with the fixation area, which encodes a further parameter of the emotional effect of the fixation area (eg attractiveness);

6 the representation of 5 wherein the fixation region is additionally given a geometric shape in the form of a thickened edge for indicating a third parameter of the emotional effect (eg activation from, for example, skin conductance changes) of the fixation region;

7 an embodiment of a visualization map, the fixation data, data on attractiveness (from facial muscle activity) and data for activation (from skin conductivity) of test persons in a superposition representation represents; and

8th an alternative embodiment of the representation of data concerning the activation (from, for example, skin conductance) of a subject.

The 1 shows the basic structure of a system for visualizing the emotional impact of a visual stimulation, whereby a visualization card is provided by means of the system.

The system provides a stimulation 1 ready with a test person 2 is stimulated. The stimulation is, for example, an image, for example an advertisement or a website, or a sequence of images (video sequence). Stimulating the subject 2 with the stimulation 1 This happens because the test person visually detects and perceives the stimulation. Stimulation may also be an audio-visual stimulation. The perception of the stimulation by the subject 2 is preferably limited to a defined time to a measurement on different subjects 2 but in principle can also be defined by the subject. Likewise, in principle, the test person can change, manipulate or terminate the visual stimulation by means of one or more input devices (for example a mouse, a keyboard, a voice key, a BCl, or the like).

During stimulation, the gaze behavior of the subject is recorded 2 in a gaze detection module 3 , Furthermore, the detection of emotional indicators, for example, by the measurement of physiological data of the subject in a detection module 4 , Different behavioral and physiological parameters (eg facial muscle activity via video analysis or electromyography) can be measured simultaneously. All data streams are synchronized in time by means of a synchronization unit 5 detected. The synchronization is realized for example via marker signals.

After the synchronized acquisition of the data, these are stored in a memory unit 6 stored and then in a module 7 processed and displayed graphically in a visualization card.

The detection of the gaze behavior of the subject 2 by means of the eye-tracking module 3 takes place via an eye-tracking device, which is known per se and which the gaze position of the subject 2 recorded and recorded in real time. The gaze behavior measurement takes place, for example, with a high-resolution high-speed camera which detects infrared light reflected by the subject's eye. The measured data are the temporal fixation of the eye, its frequency and duration of fixation. It is thus recorded where the subject looked how often, how long and in what order.

As a result of gaze behavior measurement, a particular area is defined which the subject considers at a given time and within a measured time interval. With optimal spatial precision of gaze behavior measurement, this area may be approximately punctiform or reach the size of a pixel of a computer screen. However, it can be provided that this area has a certain extent or comprises a plurality of pixels.

It is further pointed out that the gaze behavior measurement can be done with any equipment that is useful for this purpose (eg webcam-based eye-tracking).

During the stimulation, furthermore, as explained above, data on at least one further emotional effect parameter (eg physiological data) of the test person are obtained 2 in the module 4 synchronously, which may comprise a plurality of sub-modules, depending on how many emotional indicators (e.g., facial muscle activity and skin conductance) are detected.

It is provided in the considered embodiment that are measured as indicators of the emotional effect physiological data, namely on the one hand the facial muscle activity and on the other hand, the skin conductivity of the subject. The corresponding measurements give one first and a second record. In the considered embodiment, a total of three recorded data sets (gaze behavior, electromyographically detected facial muscle activity, skin conductivity) each contain a continuous data stream. This one is in the unit 5 recorded synchronized and in unity 6 saved. Subsequently, a processing and a graphical representation in a visualization card.

The fixation data, as explained above, indicates the intensity of the viewer's spatial attention with respect to different areas or stimuli of the stimulation. The subject's facial muscle activity data indicates the attractiveness of different areas or stimuli of the stimulation. The subject's skin conductance data indicates motivational activation of the subject with respect to different areas or stimuli of the stimulation.

In the following, data preparation, spatial integration and visualization are described for each of these indicators.

Indicator "attention"

Artifacts are identified and corrected in the continuous stream of recorded gaze behavior. In the corrected data stream, fixations are identified whose location is determined in the area of the stimulus (X-Y coordinates) and their duration. These steps are calculated by software. The visualization of attention is done by manipulating the visibility of the pixels of the original stimulus. The fixation duration of a fixation (and / or fixation frequency) is assigned to the pixel coordinate of the fixation site. Depending on the duration of fixation (and / or the number of fixations), a certain number of pixels thus receive a value that indicates attention. Based on the idea of foveal focus in existing parafoveal information acquisition, the surrounding pixels of this fixation site are attributed equal or gradually decreasing values (spatial filtering). The data of all fixations of a person flow into a raw visualization map of fixation times (and / or fixation frequencies). The cards of all persons are averaged.

The visualization is done by manipulating the visibility of the areas of the original stimulus. This can be done by manipulating the transparency (eg alpha blending) of the areas of a semitransparent (or opaque), colored (and / or hatched) layer overlying the original stimulus. Areas that have not received much attention have, analogously, a transparency of X in this layer (eg 50%). The more attention, ie fixation time (and / or fixation frequency) has received an area of the stimulus, the more transparent the corresponding area is shown in the colored (and / or hatched) layer overlying the original stimulus. This changes the visibility of the underlying dots / pixels of the original stimulus. The manipulation of the visibility of the original stimulus can also be achieved via distortion filters. The captured attention values determine the strength with which this filter is used. In principle, however, the strength of the visual distortion of certain areas is reduced with the intensity of the attentiveness values assigned to them.

The connection between attention value (determined from the gaze behavior, ie fixation duration and / or fixation frequency) and visibility can be defined in different ways.

According to a first variant, the relationship is self-referential: differences within a stimulus are emphasized. Within the attention card averaged over the subjects, the pixel is determined with the maximum of the attention score. This value is assigned 100% visibility and the visibility of all other pixels is referenced to this value.

According to a second variant, the relationship is criterium-related: differences between the stimuli are absolutely comparable. If there is a need to compare at least two stimuli, you can predefine which attention value corresponds to 100% visibility.

According to a third variant, the relationship is differential: differences between stimuli are emphasized. First, the attention values of all pixels of the stimulus, averaged over all subjects, become 1 from the stimulus 2 subtracted. In this differential chart, positive values indicate relatively more attention and negative values relatively less attention in stimulus 2 versus stimulus 1 , In this difference map, the pixels with the maximum value and minimum value are identified. The maximum value is assigned 100% visibility. The minimum value is assigned 0% visibility. Alternatively, the maximum and minimum values may be assigned an arbitrary visibility level (degree of manipulation of transparency of the overlay layer, degree of distorting filtering).

Alternatively, prior to averaging the attention values, the attention values of all subjects for each individual pixel may be subjected to a statistical test procedure (e.g. T-test or chi-square test). For each pixel, it is calculated with which probability the mean value of the attention values of all subjects deviates from a previously determined criterion. In the case of attention, this criterion is preferably zero (ie no attention). The probabilities of a significant deviation of the mean from the criterion can theoretically vary between 0 and 1. The higher the number, the more likely it is that the attention value determined by the gaze behavior (fixation duration and / or fixation frequency) actually deviates from zero. This method allows each pixel to be assigned a particular such value. Analogous to the procedure described above, visibility values can be assigned to these deviation probabilities and these can be visualized thereby. As described above, the attention values of one stimulus may be subtracted from those of the other stimulus to index the relative attention distribution between two stimuli. This difference map can be used to answer the question of whether one area (one pixel) of a stimulation receives more or less attention than the same area (pixels) in another stimulus. If the underlying card is such a difference card, the values of this card may also have a negative sign. In this case, this sign is considered in the representation of the probability of deviation. In this case, it is then possible to assign a specific visibility level to negative values for the visualization.

Indicator "Attractiveness"

• 0) Vorverarbeitung: 1. Die Datenströme aus der elektromyografischen Ableitung von Corrugator und Zygomaticus werden zeitlich bandpassgefiltert. 2. Artefakte werden bereinigt und 3. mittels RMS-Filter integriert und rektifiziert.
• a) Individuelle z-Standardisierung: Die Datenströme aus Corrugator und Zygomaticus werden über den gesamten Erhebungszeitraum hinweg individuell z-transformiert, damit diese dann später miteinander verrechnet werden können.
• b) Baselinebestimmung: EMG-Aktivierung sollte relativ zu einem vorhandenen Ausgangsniveau verglichen und integriert werden (Baseline). Wissenschaftlich etabliert ist es, als Baseline die Grundaktivität ohne Stimulation zu erfassen. Von den EMG-Daten wird deshalb der Median oder Mittelwert der Aktivität in einem bestimmten Zeitintervall (üblich sind 100 bis ca. 3000 ms) vor der Stimulus-Präsentation berechnet. Die Baseline wird für beide Datenströme (Corrugator und Zygomaticus) separat berechnet. Prinzipiell ist es aber auch möglich, auf die Bestimmung einer Baseline zu verzichten.
• c) Wurde eine Baseline bestimmt: Korrektur der Daten um die Baseline: Die Daten aus der kontinuierlich erfassten Aktivität der Muskeln werden im Zeitraum der Stimuluspräsentation anhand der Baseline korrigiert, beispielsweise durch Differenzbildung oder Quotientenbildung.
• d) Es erfolgt eine zeitliche und räumliche Integration der Daten: Zuerst muss definiert werden, welcher Abschnitt des Datenstroms der jeweiligen Fixation zugeordnet werden kann. Es wird der Beginn der jeweiligen Fixation, deren Lokation und Dauer identifiziert. Korrigierte EMG-Daten werden der jeweiligen Fixation wie folgt zugeordnet.
• a. Zeitlicher Versatz (Delay): das EMG-Signal unterliegt einer Trägheit. Diese wird in einer Ausführungsvariante durch einen zeitlichen Versatz korrigiert. Aktuelle Untersuchungen zeigen, dass die Attraktivität des Reizes ab ca. 500 ms im EMG sichtbar wird.
• b. Fixationsbeginn + Zeitlicher Versatz (Delay) ist gleich dem Beginn der für diese Fixation relevanten Daten ist gleich dem Integrationsstart. Das Ende der Integration wird durch die Dauer der Fixation bestimmt: Integrationsstart plus Fixationsdauer ist gleich Integrationsende. Prinzipiell kann aber auch ein kürzeres oder längeres Integrationsfenster gewählt werden.
• e) Berechnung eines Attraktivitätswertes für die Fixation: Aus den EMG-Daten können jetzt in verschiedener Art und Weise Attraktivitätswerte berechnet werden.
• a. Zwischen Integrationsstart und -ende werden die EMG-Daten von Corrugator und Zygomaticus gemittelt und erster von letzterem Wert abgezogen. Dieser Wert bildet einen singulären Wert für die Attraktivität. Es ist auch möglich, nur einen der Muskeln in einen Attraktivitätswert zu überführen.
• b. Zwischen Integrationsstart und -ende werden die EMG-Daten von Corrugator und Zygomaticus auf eine Abweichung über einen nicht näher definierten Wert (z. B. 1 Standardabweichung) vom Mittel überprüft. Sollte im Corrugator-Datenstrom ein solcher Wert festgestellt werden, bekommt diese Fixation einen Corrugatorwert von 1 (sonst 0). Gleiches gilt für Zygomaticus. Der Attraktivitätswert dieser Fixation wird durch die Subtraktion von Zygomaticuswert und Corrugatorwert berechnet (von –1 bis 1).
• c. Prinzipiell ist es möglich, nur einen der beiden Muskeln für die Erfassung der „Attraktivität eines Reizes zu verwenden. In diesem Fall besteht ein inverser Bezug zwischen Attraktivität und Corrugatoraktivität. Im Fall der Zygomatikus-Aktivität ist der Bezug positiv.

The preparation, correction of data and visualization is done by the following measures:

• 0) Preprocessing: 1. The data streams from the electromyographic derivation of Corrugator and Zygomaticus are bandpass filtered in time. 2. Artifacts are cleaned up and 3. integrated and rectified by means of RMS filter.
• a) Individual z standardization: The data streams from Corrugator and Zygomaticus are individually z-transformed over the entire survey period, so that they can then be offset against each other later.
• b) Baseline determination: EMG activation should be compared and integrated relative to an existing baseline level (baseline). Scientifically established is to record baseline activity without stimulation. Therefore, the EMG data is used to calculate the median or mean of the activity over a given time interval (typically 100 to about 3000 ms) before the stimulus presentation. The baseline is calculated separately for both data streams (Corrugator and Zygomaticus). In principle, it is also possible to dispense with the determination of a baseline.
• c) Has a baseline been determined: Correcting the data by the baseline: The data from the continuously recorded activity of the muscles are corrected in the period of the stimulus presentation based on the baseline, for example by subtraction or quotient formation.
• d) There is a temporal and spatial integration of the data: First, it must be defined which section of the data stream can be assigned to the respective fixation. It identifies the beginning of each fixation, its location and duration. Corrected EMG data are assigned to the respective fixation as follows.
• a. Time delay (delay): the EMG signal is subject to inertia. This is corrected in a variant by a temporal offset. Recent studies show that the attractiveness of the stimulus becomes visible in the EMG after about 500 ms.
• b. Beginning of fixation + Delay is equal to the beginning of the data relevant for this fixation is equal to the start of integration. The end of the integration is determined by the duration of the fixation: integration start plus fixation duration equals end of integration. In principle, however, a shorter or longer integration window can also be selected.
• e) Calculation of an attractiveness value for the fixation: The EMG data can now be used to calculate attractiveness values in various ways.
• a. Between integration start and end, the EMG data of Corrugator and Zygomaticus are averaged and subtracted from the latter value. This value is a singular value for attractiveness. It is also possible to convert only one of the muscles into an attractiveness value.
• b. Between the start and end of the integration, the EMG data from Corrugator and Zygomaticus are checked for deviation beyond an undefined value (eg 1 standard deviation). If such a value is detected in the Corrugator data stream, this fixation receives a Corrugator value of 1 (otherwise 0). The same applies to Zygomaticus. The attractiveness value of this fixation is calculated by subtracting zygomaticus value and corrugator value (from -1 to 1).
• c. In principle it is possible to use only one of the two muscles to detect the attractiveness of a stimulus. In this case, there is an inverse relationship between attractiveness and corrugator activity. In the case of zygomatic activity, the reference is positive.

• f) Graduelle Glättung (Smoothing): Da das Gehirn während einer Fixation nicht ausschließlich Informationen eines Pixels aufnimmt, sondern potenziell Zugang zu den Informationen in einem Bereich mit einer bestimmten Gradzahl um den fixierten Pixel hat (fovealer und parafovealer Bereich) wird um den Wert des jeweiligen Pixels ein radial abnehmender Verlauf implementiert. Angenommen, Pixel [X300, Y200] bekommt einen Attraktivitätswert von 1, dann bekommen die Pixel [X299, Y200], [X301, Y200], [X300, Y199] und [X300, Y201] einen etwas geringeren Attraktivitätswert. Sollte der Ursprungs-Pixel einen negativen Attraktivitätswert aufweisen, bekommen die umliegenden Pixel dagegen einen etwas höheren Wert. Insgesamt erreicht das Smoothing, dass die Attraktivitätswerte der Pixel um einen fixierten Pixel mit zunehmendem Abstand gegen Null tendieren (relativ neutraler werden, also weder attraktiv noch unattraktiv). Pixel ab einem bestimmten Abstand vom Fixationspixel werden von diesem Smoothing, und damit dem Attraktivitätswert des fixierten Ursprungspixels nicht mehr beeinflusst.
• g) Visualisierung:
• a. Die Daten aller Fixationen bzw. Fixationsbereiche einer Person werden in die rohe Visualisierungskarte für die Attraktivität integriert. Die finalisierten rohen Karten aller Testpersonen werden gemittelt. Diese gemittelte Karte wird dann visualisiert.
• b. Die unterschiedlichen Attraktivitätswerte bekommen eine bestimmte Farbe (z. B. negativ = „Cyan”, positiv = „Magenta”). Je stärker der Abstand des Attraktivitätswertes von 0, desto intensiver die Farbe. Pixel bzw. Bereiche ohne Aufmerksamkeit (ohne Fixationsdauer) sind von der Attraktivitätsvisualisierung zwangsläufig ausgenommen. Für die Visualisierung sind alternative dichotome (rot vs. grün) oder auch komplexere Farbräume (grün, gelb, rot) möglich. Alternativ kann die Attraktivität auch durch Verzerrungen der Perspektive des Grundbilds visualisiert werden. Pixel mit hoher Attraktivität erscheinen dadurch näher (vergrößert). Pixel mit geringer Attraktivität erscheinen entfernter.

Regardless of its genesis, this attractiveness value is then assigned to the pixel of the particular fixation. So while all the pixels are the raw visualization card for the attractiveness at the beginning still have a value of 0, there is then there after the integration of the data then variation.

• f) Smoothing: Since during a fixation the brain does not only pick up information from a pixel, but has potential access to the information in an area with a certain number of degrees around the fixed pixel (foveal and parafoveal area) respective pixels implemented a radially decreasing course. Assuming that pixel [X300, Y200] has an attractiveness value of 1, the pixels [X299, Y200], [X301, Y200], [X300, Y199] and [X300, Y201] will have a slightly lower attractiveness value. If the source pixel has a negative attractiveness value, the surrounding pixels get a slightly higher value. Overall, the smoothing achieves that the attractiveness values of the pixels around a fixed pixel tend towards zero with increasing distance (becoming relatively more neutral, that is, neither attractive nor unattractive). Pixels beyond a certain distance from the fixation pixel are no longer affected by this smoothing, and thus the attractiveness value of the fixed original pixel.
• g) Visualization:
• a. The data of all fixations or fixation areas of a person are integrated into the raw visualization card for attractiveness. The finalized raw cards of all test persons are averaged. This averaged map is then visualized.
• b. The different attractiveness values get a certain color (eg negative = "cyan", positive = "magenta"). The stronger the distance of the attractiveness value of 0, the more intense the color. Pixels or areas without attention (without fixation duration) are necessarily excluded from the attractiveness visualization. For the visualization alternative dichotomous (red vs. green) or even more complex color spaces (green, yellow, red) are possible. Alternatively, the attractiveness can also be visualized by distorting the perspective of the basic image. Pixels with a high degree of attractiveness thus appear closer (enlarged). Pixels with low attractiveness appear more distant.

With this indicator, too, the attractiveness values can be assigned to the visualization characteristics (eg color, or "proximity") in different ways:
Self-referential: differences within a stimulus are emphasized. Within the averaged attraction map the pixel with the maximum attractiveness is determined. This value is assigned 100% of the color gamut / sewing scale and the values of all other pixels are transformed with reference to this value.

Criteria-related: Differences between the stimuli are absolutely comparable. If there is a need to compare at least 2 stimuli, you can predefine which appeal corresponds to 100% of the color scale / sewing scale. The values of all other pixels are transformed with respect to this value.

Differential: Differences between stimuli are emphasized. First, the attractiveness values of all the pixels of the stimulus 1 from the stimulus 2 subtracted. In this difference chart, positive values indicate relatively more attractiveness and negative values relatively less attractiveness in stimulus 2 versus stimulus 1 , In this difference map, the pixels with the maximum value and minimum value are identified. The maximum value is assigned to 100% of the color gamut / sewing scale. The minimum value is assigned 0%. All values are transformed with reference to this reference.

Alternatively, the maximum and minimum values may be assigned an arbitrary color / sewing level.

Alternatively, prior to averaging the attractiveness values, the attractiveness values of all subjects for each individual pixel may be subjected to a statistical test procedure (eg t-test or Chi-square test). For each pixel, it is calculated with which probability the mean value of the attractiveness values of all subjects deviates from a previously determined criterion. In the case of attractiveness, this criterion is preferably zero (ie neutrality, neither attractive nor repulsive). Alternatively, this criterion could usefully also correspond to the mean attractiveness value of all (fixed) pixels. The probabilities of a significant deviation of the mean from the criterion can theoretically vary between 0 and 1. The higher the number, the more likely it is that the attractiveness value determined by the facial muscle activity actually deviates from zero. This method allows each pixel to be assigned a particular such value. In the case of attractiveness, a pixel may also seem rather repugnant. It may therefore be that the attractiveness value of a pixel deviates in the negative direction from zero. Before these deviation probabilities can be visualized, therefore, the sign of this deviation must first be determined. Analogous to the procedure described above, these deviation probabilities can then be assigned any color values and these can be visualized thereby.

Indicator "Activation"

• a) Zeitliche und Räumliche Integration der Daten: Die Synchronisation und Integration der dekomponierten Hautleitfähigkeitsdaten erfolgt analog zu den oben beschriebenen EMG-Daten. Zuerst muss definiert werden, welcher Bereich des Datenstroms der jeweiligen Fixation zugeordnet werden kann. Zuerst bestimmt man den Zeitpunkt des Fixationsbeginns und addiert bevorzugt einen bestimmten zeitlichen Versatz (Delay), welcher durch empirische Ableitungen bestimmbar ist. Durchgeführte Untersuchungen ergeben, dass dieser zeitliche Versatz (zwischen Fixationsbeginn und tatsächlichem fixationsbezogenem Integrationsintervall) etwa 0.5 bis 5 s beträgt. Diese Angaben basieren auf eigenen unveröffentlichten Daten aus Einzelreiz-Stimulationen. Das Integrationsende des fixationsbezogenen Integrationsfensters wird, wie oben beim EMG beschrieben, durch die Dauer der Fixation bestimmt. Prinzipiell kann aber auch ein kürzeres oder längeres Integrationsfenster gewählt werden. Der Mittelwert der Daten in diesem Integrationsfenster wird dem Pixel oder den Pixeln der jeweiligen Fixation zugewiesen. Die Daten aller Fixationen einer Person werden so in die rohe Visualisierungskarte für die Aktivierung integriert.
• b) Smoothing erfolgt wie oben beschrieben
• c) Die Datenkarten aller Testpersonen werden gemittelt. Diese gemittelte Karte wird visualisiert. Für die Visualisierung sind verschiedene Möglichkeiten vorstellbar, die in Bezug auf die 7 und 8 weiter unten beschrieben werden.
• d) Alternativ können vor Mittelung der Pixelwerte über die Probanden die ermittelten Aktivierungswerte aller Versuchspersonen für jeden einzelnen Pixel einem statistischen Prüfverfahren unterzogen werden (z. B. t-Test oder Chi-Quadrat Test). Dabei wird für jeden Pixel berechnet, mit welcher Wahrscheinlichkeit der Mittelwert der Werte aller Probanden von einem zuvor bestimmten Kriterium abweicht. Im Fall der Aktivierung ist dieses Kriterium vorzugsweise gleich Null (also Neutralität; weder attraktiv noch abstoßend). Alternativ könnte dieses Kriteriums sinnvollerweise auch dem mittleren Aktivierungswert aller (fixierten) Pixel entsprechen. Die Wahrscheinlichkeiten einer bedeutsamen Abweichung des Mittelwerts vom Kriterium können theoretisch zwischen 0 und 1 variieren. Je höher die Zahl, desto wahrscheinlicher ist es, dass der (z. B. durch die Messung des Hautleitwerts bestimmte) Aktivierungswerts tatsächlich von Null abweicht. Durch dieses Verfahren kann jedem einzelnen Pixel ein bestimmter solcher Wert zugewiesen und die Gesamtheit dann wie weiter unten beschrieben visualisiert werden.

In the person's SCR data (ie the skin conductance values) become artifacts identified and corrected, and the corrected data subjected to a decomposition analysis. This technique was published in 2010 by Benedek and Kaernbach. The output of this decomposition analysis isolates the actual sympathetic activity (bursts) underlying the skin conductance response. This is important because the skin conductance response is very sluggish (2-4 seconds) and multiple nerve impulses mingle in a skin conductivity response. It is therefore sometimes difficult to associate the skin conductance response with a particular short-term event (eg fixation). The output of the decomposition analysis, on the other hand, involves a relatively distinct operationalization of the temporally very narrowly defined nerve impulses, which can be assigned much better to short-term stimulations (fixations). In principle, the stronger the nerve impulse (neuronal firing), the greater the rash in the output of the decomposition analysis. A baseline correction of the data is not necessary because the unfolded data has an absolute zero.

• a) Temporal and spatial integration of the data: The synchronization and integration of the decomposed skin conductivity data is analogous to the EMG data described above. First, it has to be defined which area of the data stream can be assigned to the respective fixation. First, one determines the time of the beginning of the fixation and preferably adds a specific time delay (delay), which can be determined by empirical derivations. Investigations have shown that this time offset (between the beginning of fixation and the actual fixation-related integration interval) is about 0.5 to 5 s. These data are based on our own unpublished data from single stimulus stimulation. The integration end of the fixation-related integration window is determined by the duration of the fixation, as described above for the EMG. In principle, however, a shorter or longer integration window can also be selected. The mean of the data in this integration window is assigned to the pixel or pixels of the particular fixation. The data of all fixations of a person are integrated into the raw visualization card for activation.
• b) Smoothing is done as described above
• c) The data cards of all test persons are averaged. This averaged map is visualized. For the visualization different possibilities are imaginable, which in relation to the 7 and 8th will be described below.
• d) Alternatively, before averaging the pixel values over the subjects, the determined activation values of all subjects for each individual pixel may be subjected to a statistical test procedure (eg t-test or Chi-square test). For each pixel, it is calculated with which probability the mean value of the values of all subjects deviates from a previously determined criterion. In the case of activation, this criterion is preferably zero (ie neutrality, neither attractive nor repulsive). Alternatively, this criterion could usefully also correspond to the mean activation value of all (fixed) pixels. The probabilities of a significant deviation of the mean from the criterion can theoretically vary between 0 and 1. The higher the number, the more likely it is that the activation value (determined, for example, by measuring the skin conductance) actually deviates from zero. This method allows each pixel to be assigned a particular such value and then visualized as a whole, as described below.

As described above, the activation values of one stimulus can be subtracted from those of the other stimulus to index the relative activation value distribution between two stimuli. This difference map can be used to answer the question of whether one area (one pixel) of a stimulus triggers more or less activation than the same area (pixels) in another stimulus. If the underlying card is such a difference card, values of this card may also have a negative sign. In this case, this sign is considered in the representation of the probability of deviation. In this case it is then also possible to assign a specific coding level (thickness and / or color of the contour line and / or the circle) to negative values for the visualization.

The 2 to 6 clarify the inventive method using an exemplary embodiment. The 2 shows a stimulation 1 in the form of an image, which is shown schematically. For example, the image is an advertising image. The advertising agency that created the ad wants to capture its emotional impact.

A subject becomes with the stimulation 1 (hereinafter also referred to as scene) stimulated by viewing it visually for a defined period of time. The natural gaze behavior of a subject is such that it performs fixations, that is, precisely observes certain points in the image. Such fixations are measured in terms of their spatial location and duration. Despite the inevitably limited spatial precision of the measurement of the subject's gaze position, fixation is usually assigned to a specific point or pixel. Since information is usually collected in a foveal and parafoveal area, it is assumed that information is collected in a more or less extensive area takes place around the fixed pixel. One such area 11 is in the 3 shown. After looking at the area 11 for a certain time the subject will fix another area, etc.

In the following, only one such area is considered to illustrate the method according to the invention. For the purposes of the invention, however, (usually) an evaluation of the gaze behavior of the test person can take place over the entire exposure time and over the entire extent of the stimulus.

The subject looks at the area 11 , hereinafter also referred to as fixation area, for a certain period of time (ie, during a certain time interval) and with a certain frequency. The length of the time interval, ie the duration of the viewing (or the frequency of viewing) of the area 11 is reflected by the extent of visibility of the underlying stimulus. For example, a long (and / or frequent) viewing of a particular area results 11 This area may appear more visible than an area due to greater transparency of a superposed dark layer 11 which is only considered for a short time or not at all. The transparency of the overlying layer 12 is in accordance with the 4 also shown.

Looking at the area 11 the scene 1 by the subject is in fact a stimulation with an attractiveness to be measured. The attractiveness is measured by the acquisition of physiological parameters concerning facial muscle activity, as explained above. The measured attractiveness is integrated spatially and temporally in relation to the fixation and displayed color-coded. For example, a high attractiveness is represented by the color magenta and a low attractiveness by the color cyan. The color 13 is in accordance with the 5 also shown (here: by hatching).

The observation of the area or scene component 11 by the subject is further associated with a certain motivational activation. This is, as explained above, for example, measured by measuring physiological parameters with respect to the skin conductance of the subject. The motivational activation is represented by a geometric form. In the embodiment of 6 is this geometric shape by the thickness of the peripheral edge 14 indexed the area 11 surrounds. The larger the thickness of the peripheral edge 14 The greater the degree of motivational activation, for example.

It should be noted that the indicators of emotional impact, which are evaluated in terms of attractiveness and activation (especially physiological data on facial muscle activity and skin conductance), with a time lag in the area 11 and the associated time interval. This takes into account a delayed emotional reaction to the viewing of the area 11 connected stimulus. Thus, the data relating to the respective emotional indicator come from a time interval that at the time interval that the fixation data of the area considered 11 contains, is time-shifted.

The indicators 12 . 13 . 14 Attention, attractiveness and activation are in the 2 to 4 only for an area or a scene component of a scene 1 shown. The representation is made in accordance with the present system and method for all areas of the scene 1 looking at the subject.

The result of such a spatial evaluation of the complete stimulus shows in an alternative embodiment the 7 , In the 7 For the complete stimulus (the whole picture 1) the indicators attention (transparency coded or brightness coded), attractiveness (color coded) and activation (coded by circles) are indicated.

The visualization of the emotional parameter "activation" takes place via the marking of the strongest activating regions by concentric circles. This visualization form is a simplification of the contour representation, which in view of the 8th will be explained. From areas within a defined isovalent boundary (eg, 80% of the maximum value of an activation map averaged over the subjects), the geometric centroid is determined. This is defined as the center of a circle. The area of the circle is determined by the area of the area within the underlying isovalent contour. The threshold value (eg 80% of the maximum value of an activation card averaged over the subjects) of this contour determines the thickness (and / or color) of the circle.

A visualization map is created 10 , which in the considered embodiment, but not necessarily, the image 1 is superimposed. The visualization card 10 allows the emotional effect of stimulation 1 to grasp and comprehend at a glance.

Numerical values can also be assigned to the indicators, with a high numerical value signifying a high degree of attractiveness and / or high activation. These numerical values can be compared with different images to determine if the emotional effect of a distinguishes statistically significant between two general images.

8th Figure 9 shows an alternative embodiment in which the data indicative of motivational activation, ie, in the described embodiment, the physiological data with regard to skin conductivity, are not geometric shapes (circles) but instead contour lines 15 are shown. For this purpose, the maximum is determined within the activation card averaged over the test persons. On the basis of this value, one (or more) criterion threshold (s) is determined: eg 80% (60%, 40%, etc.) of the maximum value. Pixels with these thresholds are connected within the visualization map in the sense of isovalent boundaries. The value of this threshold can be indexed by the thickness and / or color of the line. The viewer can thus identify areas that exceed a certain activation threshold. The definition of these thresholds can be self-related, criterion-related or differential as described above.

The invention is not limited in its embodiment to the embodiments shown above, which are to be understood only as examples. For example, other behavioral predictive indicators of emotional effectiveness (eg, other physiological and / or behavioral parameters) may be measured and otherwise presented in a visualization map.

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited patent literature

• US 2008/0965468 A1 [0009]

Claims (17)

Method for visualizing the emotional effect of visual stimulation ( 1 ) comprising the steps of: stimulating a subject ( 2 ) with the visual stimulation ( 1 ), - measuring the gaze behavior of the test person ( 2 ) during the stimulation, - measuring at least one indicator of the emotional effect on the subject ( 2 ) in synchronism with the stimulation and the detection of the gaze behavior, - determination of fixation data from the measurement of the gaze behavior, which indicate which areas ( 11 ) of stimulation by the subject ( 2 ) for which time intervals and / or how often have been fixed (fixation areas), - assigning to the fixation areas determined in this way ( 11 ) of data concerning the at least one indicator of the emotional effect, and - representation of the fixation areas ( 11 ) and the respectively associated data concerning the at least one indicator of the emotional effect in a visualization map ( 10 ). Method according to claim 1, characterized in that the method is applied to a plurality of test persons ( 2 ) and their measured data are averaged. A method according to claim 1 or 2, characterized in that the fixation data displayed brightness-coded or visualized in another way in terms of their visibility. Method according to one of the preceding claims, characterized in that the data relating to the at least one indicator of emotional effect are color-coded and displayed relative to the fixation. Method according to one of the preceding claims, characterized in that as the first indicator of the emotional effect the facial muscle activity of the subject ( 2 ) and at least one other indicator of the emotional impact of the subject ( 2 ) are measured during stimulation, wherein first data regarding the facial muscle activity and second data concerning the at least one further indicator of the emotional effect are measured, and wherein the first data is color-coded and the second data is represented by geometric shapes or contour lines. Method according to one of the preceding claims, characterized in that the fixation areas ( 11 ) and the respectively associated data concerning the at least one indicator of the emotional effect in the visualization map ( 10 ) are superimposed. Method according to one of the preceding claims, characterized in that the assignment of the data concerning the at least one indicator of the emotional effect to the fixation areas ( 11 ) takes place with a time delay in the sense that data are assigned to a contemplated fixation area that originate from a time interval which is time-shifted to the time interval in which the observed fixation area was fixed by the test subject with his gaze. A method according to claim 7, characterized in that the time shift between 0.01 s and 10 s, in particular in the range between 0.4 s and 2 s, in particular in the range between 0.5 s and 1 s. Method according to one of the preceding claims, characterized in that the stimulation is a 2-dimensional or 3-dimensional (possibly holographic representation of a) individual scene or sequence of individual scenes. Method according to one of the preceding claims, characterized in that the data streams of the fixation data and the data relating to the at least one indicator of emotional effect are recorded synchronized in time. Method according to one of the preceding claims, characterized in that the fixation data and / or the data concerning the at least one indicator of emotional effect are smoothed around a considered fixation area. Method according to one of the preceding claims, characterized in that the one considered fixation area ( 11 ) spatially and temporally integrated data concerning the at least one indicator of emotional effect and in the visualization map ( 10 ) are spatially and temporally integrated. Method according to one of the preceding claims, characterized in that the at least one measured indicator of the emotional effect is a physiological, a behavioral or a psychological parameter, for example a facial muscle activity, a skin conductivity, a pupil dilation, an electrocardiogram, a heart rate variability, by means of an electroencephalography Device measured parameter, a measured by a device for functional magnetic resonance imaging parameter, heart rate and / or skin temperature, a head tracking, a motion tracking, an approach and Avoidance behavior, other muscle contractions, and / or open behavior of the subject. Method according to one of the preceding claims, characterized in that at least two indicators of the emotional effect on the test person are measured and displayed synchronously with the stimulation and the detection of the gaze behavior. System for visualizing the emotional impact of visual stimulation, comprising: - means for providing visual stimulation ( 1 ) for a subject ( 2 ), - Medium ( 3 ) for measuring the gaze behavior of the subject ( 2 ) during stimulation of the subject, - means ( 4 ) for measuring at least one indicator of the emotional effect on the subject ( 2 ) during stimulation of the subject, - means ( 7 ) for determining fixation data from the measurement of the gaze behavior, which indicate which areas ( 11 ) of the stimulation the subject ( 2 ) for which time intervals and / or how often fixed (fixation areas), - means ( 7 ) for assigning to the fixation areas determined in this way ( 11 ) of data concerning the at least one indicator of emotional impact, and - means ( 7 ) for displaying the fixation areas ( 11 ) and the respectively associated data concerning the at least one indicator of the emotional effect in a visualization map ( 10 ). Visualization card ( 10 ) the emotional effect of a visual stimulation, produced by the method according to claim 1. A computer program with software code for performing the steps of claim 1 when the computer program is run on a computer.

Images