Note : Les descriptions sont présentées dans la langue officielle dans laquelle elles ont été soumises.
CA 02248672 1998-09-10
WO 97/38624 PCT/NL97/00188
MEASURING AND PROCESSING DATA IN REACTION TO STIMULI
The invention relates to a method and device for col-
lecting and processing stimuli data and measuring the reac-
tions to said data.
Such a method is known from US-A-5,226,177. With this
method, the object is to find out how a number of respon-
dents react to e.g. advertisement illustrations and slogans,
tv and cinema commercials, images of persons, logos and
other things presented to them. With the known method, one
can employ, among other things, answering through push
buttons. This can lead to deviations and measuring errors,
e.g. by incorrect operation of the push buttons, which can
be the cause of incorrect final results. Such errors in
final results can also be caused by the fact that respon-
dents feel hindered to give their true opinions, and give an
opinion best suitable to the buttons.
Therefore, the object of the invention is to provide an
improved method, by which in a short time, e.g. within one
day, a measurement with one hundred or more respondents and
with several hundreds of stimuli can be executed, said
measurement being applicable in several places, also on an
international scale, simultaneously or otherwise. According
to the invention, such a method is characterized in that the
physiological reactions by respondents to the stimuli pre-
sented, or to components of said stimuli, are measured
automatically and are recorded in one or more computer
systems and that de recorded data, stimuli data, combined
with respondent data or otherwise, are automatically proces-
sed, e.g. in central units, to interpretable results, said
results stored in a database, e.g. a relational database,
being accessible to anyone on certain conditions.
The invention relates to a method, as well as a device
for massively collecting, processing and making available
stimuli performance data and random check survey data within
a short time, by massively presenting stimuli to large
amounts of respondents, and measuring and recording physio-
logical reactions to said stimuli.
CA 02248672 2005-10-13
2
According to a further broad aspect of the present
invention there is provided a method of collecting and
processing stimuli data and random check survey data and
presenting stimuli to respondents. The method comprises the
steps of presenting the stimuli to respondents and measuring
and recording the physiological reactions to the presented
stimuli with several systems. The respondents, the stimuli,
and components of the systems move in relation to each other
and are brought into spatial synchronization with each other
per unit time, regardless of dimensions and movements of the
respondents, the stimuli, and the components of the systems.
The method also comprises presenting the stimuli in large
quantities and processing the data to interpretable results,
and storing the processed data in a searchable database.
A preferred embodiment of the present invention will
now be described with reference to the accompanying drawings
in which:
FIGURE 1 is an example of a stimuli; and
FIGURES 2 and 3 are block diagrams illustrating
implementations of the method of the present invention.
The stimuli concern e.g. visual stimuli, including
printed matter, products, illustrations, photos, texts,
instructions, manuals, etc., in printed media, including
magazines, newspapers, specialist journals, brochures,
flyers, free local papers, DM, books, guides, etc. But also
stimuli such as e.g. TV-stimuli, productions, spots and/or
packages, shop formulas, design, art, concepts, apparatus,
models (e.g. car models), reality and photo product concepts,
as well as presentation by e.g. projection on screens,
including TV-screens.
With the method according to the invention, it is
possible to perform qualitative research on a quantitative
scale. It unites the high degree of reliability of the
CA 02248672 2005-10-13
2a
quantitative research with the great variety in subjects and
depth of the qualitative research.
The device to be used with the method according to the
invention comprises a combination of apparatus and subsystems
organizationally enabling measurement of the reactions of
e.g. more than one hundred persons a day to e.g. several
hundreds of stimuli. With the data obtained by said
apparatus, for the first time it is possible to produce
accurate calculations as regards to qualitative information.
At the same time, for the first time it is possible to
mutually compare results of calculations and thereby
obtaining judgments and insights.
The possibility of confronting e.g. more than one
hundred respondents to several hundreds of stimuli in the
is same way on one day, recording both the visual and other
physiological reactions to said stimuli, processing said
reactions to clear measurement reports and makirig said
reports available for interested persons within a few hours,
in any case within one day, is a breakthrough in relation to
all existing methods, applied technologies and devices.
Persons interested in the measurement reports are e.g.
companies advertising, marketing officials, design, film and
television production, product development, media
proprietors, etc. Persons can take an interest in the stimuli
data, to the random test data and/or to the data concerning
the reactions of the random test on the stimuli presented.
CA 02248672 1998-09-10
WO 97/38624 PCT/NL97/00188
3
For example, with the use of advertising in marketing
activities, everything is about the effectivity thereof. It
is of the utmost importance, to find out to what extent an
advertisement message is absorbed, and if so, how this can
be maximised. Here, the detailed data of the measurement
reports play an important role. A lot of information becomes
available to the commercials designers, by which e.g. short-
comings can be removed.
For example, just like with the known method, the
stimuli can be presented on screens, e.g. television
screens. Presenting the stimuli to the respondent can take
place by means of a computer-controlled program for reprodu-
cing on displays stationary or moving images, combined with
sound or otherwise, in which the program is adjusted accor-
ding to items and for the purpose of presenting stimuli in
certain sequences, dependent on demographic data and/or the
recorded physiological reactions of a respondent during the
measurement.
Here, the computer systems provide for coordination
between the presentation, the recording and the adjustment
of the presentation program employing a common time base.
However, other than with the known method, with the
method according to the invention, particularly the presen-
tation of the stimuli to the respondents can also take place
in the form of printed media, e.g. magazines and/or newspa-
pers. This way of presentation corresponds to the natural
situation. In that situation, the computer systems provide
for recording and identification of the stimuli, the presen-
tation thereof occurring as a consequence of the opening of
the pages by the respondent. At the same time, the computer
systems provide for time and duration recording per presen-
ted pages (identified by the computers).
One of the important physiological reactions concerns
the direction of the centre of the eye. Therein, it is
typical, that as the respondent is allowed more natural
freedom, the accurate measurement values are more difficult
to obtain.
In order to be able to determine the position a respon-
dent focusses the centre of his eye on, other methods apply
CA 02248672 1998-09-10
WO 97/38624 PCT/NL97/00188
4
a number of different procedures. The method mentioned
earlier is characterized by a data recording unit which is
positioned on the head of a respondent (headset or "glas-
ses") and is secured by means of a clamping band. Here, the
condition is that the "glasses" can maintain an unchanged
position in relation to the head.
As regard to the images, the output of these "glasses"
is approximately in adjustment, but movement of the glasses
cannot be prevented. Also, weariness and resistance general-
ly occur within 10 minutes. The output is mostly just one
videotape with the measurement data of one or some respon-
dents, whose images should be processed and interpreted
mainly frame by frame. Furthermore, the results can be
affected.
The known eye movement recording systems operate on the
basis of fixed references. The stimulus is fixed to the
sensor, or the glasses have been fixed on the head, or the
head is fixed in relation to the sensor, in any case, there
are always geographical references.
With the method and the system according to the inven-
tion, no fixed reference is applied. Instead, a clock time,
e.g. atomic clock, is applied for all moving parts. All
parts, including the respondent's head, the eyes, the radia-
tion sources, the reflections at the cornea, the mirrors,
the sensors, the arrangement and the stimuli, are allowed to
move in relation to each other within the room. Per unit of
time, all moving parts are brought into synchronization
adjustment in relation to each other (adjustment: a prese-
lected arrangement of the parts). Time is the fixed referen-
ce, without geographical reference. In this way, a fully
natural freedom for the respondents is realized for the
first time. The head can freely move within wide limits.
There is no fastening of parts to the head and the natural
freedom of head movement, e.g. sitting on a chair, is not
hindered or even affected. In order to be able to realize
this, among other things, the positions of the head, the
eyes, the pupils, the pupil centres, the reflections at the
cornea, the position of the eyes in the eye sockets, as well
as the position of the stimuli in relation to the eyes, are
CA 02248672 1998-09-10
WO 97/38624 PCT/NL97/00188
synchronously determined at every instant of measurement. To
that end, the device is provided with accurate sensors for
being able to find the positions in time and follow them and
for being able to use this information for bringing the
5 parts in adjustment, with the result, that the head and the
stimulus can move freely and yet the eye position on the
stimulus can be determined accurately. This method can be
applied both in its entirety and partly.
The method according to the invention is fully automa-
ted. Physiological reactions of respondents, such as eye
activity, including movements of the eye, viewing direction,
pupil size, blinking frequency, position, orientation and
movements of the head, eye-stimulus distance, as well as
changes thereof in time, etc., are measured in real time,
digitized, recorded and are subsequently, without manual
functions or personal intervention, processed to results
on-line. Due to this, respondents, operators and/or third
parties basically don't have the opportunity to influence
the results of the measurements.
When compiling the program to be performed, one can
already bear the person in mind, e.g. preferences and/or
interests of the respondent. Variables such as sex, age,
demographical data and social situation are preferably
recorded for each respondent individually.
Further, according to the invention it can be provided
for, that preceding the presentation of the interactively
selected stimuli or otherwise, the absorbing capacity, the
tempo, the reading ability, the respondent's interest for
product categories, and/or all other conceivable personal
characteristics, are determined and that by way of those,
through printed matter or e.g. through models in 3D, the
program on the screen is adjusted to the respondent concer-
ned. To that end, the determined personal characteristics
can e.g. be recorded through a chip card. By way of the data
on the chip card, e.g. the program items can be compiled.
Like with the known method, there is the possibility of
carrying out a survey in a number of places or locations at
a distance from each other. An extension of the method
provides for the possibility of putting together the measu-
CA 02248672 1998-09-10
WO 97/38624 PCT/NL97/00188
6
rement data of a number of individual systems, geographical-
ly separated or otherwise, in a central processing unit and
processing them to results as a whole.
By measuring certain physiological reactions of respon-
dents, it is possible to not only measure the reaction to
one stimulus in its entirety, but also the reaction to
certain components thereof.
Apart from the physiological reactions already mentio-
ned, furthermore, according to the invention one or more of
the following or other physiological reactions of respon-
dents can be measured and recorded:
- physiognomy, such as the positions of the corners of
the mouth, of the eyebrows, the tensions on the facial
muscles, etc.;
- blood pressure;
- heartbeat;
- breathing;
- muscle tensions;
- skin temperature;
- skin resistance;
- brain waves;
- blood flow throug a part of the body, such as e.g. the
earlobe, etc.;
- hand movements;
- the voice;
- etc.
On finding some of these last-mentioned reactions,
contrary to the reactions mentioned earlier, one can parti-
cularly act in such a way, that sensors, electrodes or other
means to be fixed to the body might have to be employed.
Also, it might be that certain measured reactions can only
relate to the overall impression of stimuli shown.
With measuring eye activities, one can operate particu-
larly in such a way, that one or two eyes are radiated at
with invisible or unnoticable reflecting radiation and
recordings are made with one or more recording elements
connected to one or more computer systems, e.g. position
sensors, eye and pupil detection sensors, other sensors or
cameras, by which e.g. at the same time the eye convergence
CA 02248672 1998-09-10
WO 97/38624 PCT/NL97/00188
7
process can be recorded. Here, radiation can be of a conti-
nuous nature but e.g. intermittent too, e.g. adjusted to the
measuring instants of the sensors. The radiation can origi-
nate from optical radiation sources, such as those of visi-
ble light, IR and UV optical radiation, but also from all
conceivable sources of radiation of another type, such as
those of ultrasound, radar and X-ray.
One can also act in such a way, that for exposure or
recording, optical elements are employed, mirrors having
certain transmission and reflection properties, for example.
Larger mirrors and application of several image-following
sensors enables all occurring head movements e.g. suitable
with a sitting posture, to be tolerated without loss of an
accurate indication about what the centre(s) of the eye(s)
is/are focussed on. Owing to this, at the same time, the
admissible dimensions of the stimuli become virtually unli-
mited and the accurate indication will also be maintained if
larger stimuli are presented, moving and/or rotating or
otherwise. With extensive stimuli, such as parts of shops,
such as shop shelves, for example, we no longer speak of
presentation.
For measuring the eye viewing direction, e.g. by means
of a moving source of radiation, the eyes can be radiated at
and the place and orientation of the pupils can be determi-
ned. Furthermore, the movement of the pupil, or its centre,
can be measured in relation to the image of the radiation
source at the cornea. From the relative positions of the
radiation source images in relation to one or two pupils in
the recorded images, the eye viewing direction can be deter-
mined.
When using radiation or light sources, dark pupils in
relation to illuminated irises are achieved. When using
radiation or light sources being positioned coaxially in
relation to the camera, pupils brightly illuminated in
weakly illuminated surroundings are achieved. Both possibi-
lities and variations thereof, and other possibilities can
be used as desired, separately as well as combined.
The eye viewing direction, position of the head, and
viewing position in a plane, can only be determined accura-
CA 02248672 1998-09-10
WO 97/38624 PCT/NL97/00188
8
tely if the eyes, the pupils, the sensors, the exposures and
possibly other, optical or not, components are brought in
very accurate adjustment to each other in time, or, conti-
nuously or not, are kept in adjustment to each other. The
system according to the invention provides this necessity by
means of dynamic calibrations, measuring the posture and
position of head, pupils, eyes with light reflections and
stimulus and spatial following thereof. In order to realize
this, a number of known and other means are applied, bots
separately and in combination.
The measurements of position, orientation and movements
of the head, as well as the distance between head, eyes and
stimuli, and the changes therein in time, provide less
detailed, yet not less important information about the
reactions of respondents to presented stimuli. For example,
having a stimuli identification and positioning system
record how the process of persons reading, leafing through
magazines, newspapers or other printed matter, such as DM,
goes. Such as: which pages were opened, for how long did the
respondent look to the left, the right, the bottom and the
top? When looking carefully at stimuli, including the above,
but also other stimuli such as packages etc., one often
tends to bend forward. A reduction of the distance between
head and stimulus can go together with giving attention. The
reverse also holds good: An increase of the distance between
head and stimulus can go together with slackening of attent-
ion. These examples show that in this way, a large amount of
information about the behaviour with objects and printed
matter can be obtained without the application of eye move-
ment registration.
For the sake of determining individual facial features,
and the'ability to correct them, a calibration can be provi-
ded, e.g. by asking respondents to look a certain locations
in a plane. It has appeared, that on looking under orders to
certain locations in a plane, the eye is able to take up an
angular position, which e.g. can result in deviations that
can have the size of a circle having a diameter of some
tenths of millimetres, e.g. 50 mm, taken across the usual
viewing or reading distance, which is so inaccurate, that
CA 02248672 1998-09-10
WO 97/38624 PCT/NL97/00188
9
therefore, the subsequent measurements get indications of
fixation locations that can be situated at a distance of
+ 50 mm or - 50 mm from the actual fixed object. According
to the invention, one acts in such a way, that for determi-
ning the calibration, one uses locations with figures con-
sisting of a number of characters being different per group,
e.g. dots or rectangles, the number of which should be
counted aloud by the respondent. One can only count with the
fovea part of the eye. Since the respondent, in order to be
able to count, must aim the fovea centralis accurately, a
careful calibration can be performed while he is counting.
Here, it should be observed, that the retinal part of the
eye is always used, whereas it doesn't require an exact
position of fixation. It can observe many things simultane-
ously, but none of it accurately. It is misleading for use
with calibration purposes: while looking at one of the outer
letters of Figure 1, one can also see the other two. Howe-
ver, in order to be able to count the number of points in
the circles behind the letters of Figure 1, the fovea cen-
tralis must be aimed at them. Couting aloud indicates the
moment at which this happens.
For presenting the stimuli, the system is equipped with
one or more stimuli presentation units, e.g. displays,
including video screens, one or more reading tables and/or
other stimuli presentation means. A reading table could be
provided with lighting and/or a clamping device for positio-
ning and fixing printed media. For clamping printed media, a
number of known methods and other methods can be employed,
both individually and combined.
For measurement and registration of the reactions to
the presented stimuli, the systeem is e.g. equipped with one
or more measuring members and/or recording elements, e.g.
sensors, including cameras; buttons, including push buttons,
with interactive reaction or otherwise; members for touch
screen answers; and/or other or further measuring members
e.g. for measurement and registration of blood pressure,
brain waves, etc.
According to the invention, the system for application
with the method described above will comprise central compu-
CA 02248672 1998-09-10
WO 97/38624 PCT/NL97/00188
ters e.g. being in communication with a number of substanti-
ally identical measurement and registration devices arranged
at spaced apart locations, each device consisting of compu-
ter units and measuring members connected to said computer
5 units, for measuring the physiological reactions of the
respondents to the presented stimuli. The computer units and
the central computers dispose of a common time base, or of
relatable or synchronizable time bases, as a consequence of
which all measurement results of each period and each loca-
10 tion can be compared and combined. Due to that, the measure-
ments of each of the measurement units performed in interval
of time can be cumulated legally, which is necessary for a
random check n = 100, in which e.g. 3 papers are measured in
6 hours.
The raw data of the measurements comprise, among other
things, the respondent reactions and the stimuli data cou-
pled to them by means of the time axis. From one or more
locally arranged devices, the raw data are transmitted to
the central processing units such as main computers, and are
subsequently processed to data e.g. per measurement, per
stimulus, per stimulus item and per respondent. The resul-
ting data are stored in a database, e.g. a relational data-
base, e.g. per measurement, per stimulus, per stimulus item
and per respondent. It is is also possible to store refe-
rences to the data in the database, instead of storing the
actual data. The references indicate where the actual data
are stored and can be found. This method is particularly
preferred in case of large files and/or a fast increase in
volume of the files.
In addition, among other things, additional data are
collected, classified and stored in the database. Advertise-
ments are e.g. stored in classes. Classes concern e.g.: in
which media the advertisements were placed, the release
dates, page numbers, dimensions and use of colours. Yet also
the positions in advertisements, indicated by means of
boxes, polygons, contours and/or other shapes, where the
advertisement elements are situated, such as: the logo, the
products, the texts, the headlines, etc.
For example, sex and age of respondents are stored. But
CA 02248672 2005-10-13
11
also the demographic data, education and social data.
Furthermore, the personal interests and the physical and
psychological capacities, including sight, reading and
absorbing capacities, spectacle corrections as well as the
tempo of absorption and processing, etc.
Through dialogue units, the database is accessible, at
a distance and under certain conditions or otherwise, and the
database can be 'interrogated' about everything conceivable
which is related to, or might be related to, the data stored
in the database and, per subject, combinations of data. By
means of software programs, the results of that can be
specified into information providing insight, that can be
presented as measurement reports.
The measurement reports can contain information about
one of the stimuli presented to respondents, or about a
selected number of stimuli. The reports contain e.g.
information about publicity of certain brands; about certain
activities; about advertisements placed in certain media, or
about certain product categories. The measurement reports can
also concern data of the random check survey, or subgroups
thereof. For example, the reactions to stimuli by men, women,
or men over 30, etc. Further, the measurement reports can
concern any conceivable combination of data.
As shown in FIG. 2, a stimuli presentation unit 1 and a
respondent reaction unit 2 are controlled by a script unit,
or script generator 3. On the one hand, the script unit
provides for presentation of the stimuli and on the other for
coordination of the measured reactions of the respondent to
those, and, subsequently, for interactive adjustment of the
further progress of the script.
The stimuli are presented on a screen 4. In connection
with the respondent's ability to ask questions about the
stimuli shown and his ability to answer, the screen can be
equipped with facilities for those purposes, e.g. window
and/or touch screen facilities. Audio stimuli, including
CA 02248672 1998-09-10
WO 97/38624 PCT/NL97/00188
12
acoustic signals, but e.g. also questions and/of as-
signments, are reproduced through a signal giver, a loudspe-
aker 5, for example.
The respondent reaction unit receives, measures and
registers a respondent's reactions, consisting of voice
reactions through microphone 6, pressure or touch reactions
through push button, touch screen or all other conceivable
means by which respondents can express their reactions 7,
eye reactions symbolically indicated by eye 8, and physiolo-
gical reactions by means of physiological sensors 9.
The reactions to stimuli presentation interactively
affect e.g. the script unit 3, so that adjustments to the
further progress of the script during a program are always
possible.
From one or more locally arranged devices, the raw data
of a measurement, obtained from the respondent reaction
unit, as well as the "stimuli history", that is, the stimuli
program shown e.g. under interactive action, are sent fur-
ther to central processing units 10, such as main computers,
so that said raw data can be processed to data e.g. per
stimulus and per respondent. Said data are stored in a
database 11. By means of one or more dialogue units 12, the
relational database, at a distance and under certain condi-
tions or otherwise, are 'interrogated" about everything
conceivable which is related to, or might be related to, the
data stored in the database and, per subject, combinations
of data. By means of software programs, the results of that
can be specified into information providing insight, that
can be presented on screens or in the form of printed re-
ports.
Figure 3 shows, as an example of another embodiment of
the method according to the invention, a flow diagram in
which printed stimuli are presented, positioned on one or
more reading tables. The stimuli presentation and registra-
tion unit comprises one or more reading tables on which the
print stimuli, e.g. in the form of magazines and papers, are
positioned, and fixed or otherwise. For example, by opening
or not opening the pages, respondents choose the stimuli
themselves. All stimuli presented by opening the pages are
CA 02248672 1998-09-10
WO 97/38624 PCTINL97/00188
13
registered and, contrary to other methods, automatically
identified. To that end, the system consists of one or more
sensors by which characteristics of stimuli can be registe-
red and processed and are kept available for comparison. The
system works both on the identification of a stimulus in its
entirety and on partly invisible stimuli, including stimuli
largely covered. For example, a partly opened magazine or
paper will already be sufficient to be able to determine the
number of that page with certainty. Rotations up to 1800 are
tolerated. The processing is fast, yet not always real time,
which is not always necessary. There are a number of known
and other identification methods, which can be employed, in
combination or otherwise, for identifying stimuli.
Furthermore, with the method according to the invention
it is possible to determine the position and/or orientation
and/or distortion of the stimulus per measuring instance
and, either simultaneously or later, compensating for them.
Distortion of stimuli can be the consequence of e.g. magazi-
ne or paper pages bulging. In this way, it is possible to
register manipulations of and with stimuli by respondents,
such as with printed matter the way of holding and/or moving
pages. Yet also manipulation of objects, e.g. turning ob-
jects for the purpose of looking at them in several views.
Preferably, the measuring situation is such, that during the
instances of recording, the stimuli are not or only slightly
moving.
With application of the method according to the inven-
tion, the data relating to the head posture, head position,
position of the eyes, eye position in the eye sockets, pupil
position and viewing direction will be related to above-
mentioned stimuli data. Owing to that, for the first time it
is possible to determine, dynamically and for each measuring
instance, which stimuli, and which parts thereof, are visi-
ble, which part thereof is actually in the respondent's
visual range, and, in particular, to accurately determine
for each instance on what stimulus part the centre of the
eye, the fovea centralis, of the respondent is focussing.
In other words, the position and orientation of the
stimuli in relation to the position and orientation of the
CA 02248672 1998-09-10
WO 97/38624 PCT/NL97/00188
14
head together with the "visual field" of the respondent and
the visibility of the stimuli parts. The position and orien-
tation of the stimuli in relation to the positions and
orientations of the eyes together with the visibility of the
stimuli parts provide the parts of the stimuli which will be
depicted on the "fovea centralis". All "determinations" of
all parts concerned should be made for each measuring in-
stance in such a way, that the information can be used for
bringing said parts in register.
In addition to registration and identification of the
stimuli, the starting time and the length of time are deter-
mined, during which e.g. pages of magazines are lying open
and thus stimuli can be visible. Depending on the possible
exceeding of a, e.g. preset, maximum length of time to be
spent per stimulus, per page, per double page or per magazi-
ne, a signal as a reminder can be given through a signal
giver, e.g. a loudspeaker 5.
The method according to the invention also provides for
identification of the behaviour characteristic of reading.
Depending on the possible exceeding of a, e.g. preset,
maximum length of time to be spent reading, including e.g.
reading headlines of advertisements or editorial texts in
magazines, likewise, a reminder can be given through the
signal.
For the rest, the flow according to Figure 3 corres-
ponds to the one described with Figure 2.
By way of example, step by step a method will be des-
cribed, which can be followed together with the process and
system according to the invention for determining the posi-
tion on a stimulus on which the centre of the eye of a
respondent is focussed.
1.' Determining the presence of the head by means of a
multiple sensor device composed of one or more, e.g. image
producing, sensors, including e.g. one or more cameras.
2. Determining the 3-dimensional position and orienta-
tion of the head, e.g. with the help of the device as des-
cribed at 1.
3. Determining the most likely position of the eyes in
the head, of the centre of the eye and the point of reflec-
CA 02248672 1998-09-10
WO 97/38624 PCT/NL97/00188
tion on the cornea, e.g. with the help of the device as des-
cribed at 1.
4. Controlling the geometrical range of one or more
eye detection sensors, such as e.g. adjusting the image
5 field of one or more image forming sensors, such as cameras,
by one or more motor drives, to the position of the eyes and
subsequently setting the sensors in relation to the eye
distance, such as focussing when using depicting optics,
followed by checking the set geometrical range and, if
10 necessary, correcting and adjusting them.
Preferably, the eye detection sensors are separate sensors
being able to operate parallel to, coupled to, but e.g. also
more or less indenpendent of, the sensors described at 1.
For irradiation of the eyes, the eye detection sensor can be
15 equipped with one or more radiation sources preferably
having invisible beams of rays. By coupling the sources to
the sensors mechanically, for example, the geometrical range
of the assembly can be controlled synchronously. The advan-
tage thereof is, among other things, that there is directio-
nal irradiation and that a higher radiation intensity on the
eyes is achieved with less energy and that relatively small
radiating angles of the sources suffice for achieving a
large spatial range.
5. Determining that the eyes are within reach of the
eye detection sensors.
6. Determination of the momentary characteristics of
eyes, eyelids, corneas and pupils, such as e.g. the degree
to which the eyelids are open, or to what extent the corneas
or the pupils are covered by the eyelids; the pupil sizes
and the contrasts between iris and pupil.
7. Determination of the transitions between pupil and
iris.
8. Reconstructing the transition curves between pupil
and iris, determining the degree of roundness of the pupils
and, e.g. in case of elliptical shape, caused by natural
distortion or perspective, for example, determination of the
ellipse axes and the spatial orientations and positions of
said axes.
These reconstructions and determinations serve, among other
CA 02248672 1998-09-10
WO 97/38624 PCT/NL97/00188
16
things, for being able to accurately determine the pupil
centre, despite a perspective distortion as a consequence of
the sensor-pupil geometry.
9. Determination of the pupil centre.
10. Determination of the centre of the positions of the
reflections of the radiation sources on the corneas, e.g. by
generating histogram distributions of e.g. the radiation
intensities of the reflections in different directions.
11. Determination of the distortions of e.g. an image
of said reflections in relation to the spatial profiles of
the emission sources as a consequence of characteristics of
the eyes, including aqueous humour and/or the eye distances
in relation to the eye detection sensors.
12. Calculating the viewing directions in the room.
13. Correcting the calculated viewing directions with
the previously measured individual parameters of the eyes,
including e.g. shift, scale and similar and other (non-)
linear corrections.
14. Correcting the spatial eye positions determined at
3.
15. Correcting the geometry and/or optics of the arran-
gement.
16. Bringing the multiple corrected viewing directions
in register with the 3-dimensional position and orientation
of the positioning planes, such as tabletops of reading
tables, of the stimuli to be presented.
17. Calculating the fixation positions on the positio-
ning planes of the stimuli to be presented.
18. Measurement and calculation of the 3-dimensional
orientation, position and situation of a presented stimulus,
e.g. a printed stimulus, including e.g. a page of a magazine
or paper, or an object, in relation to the positioning
planes.
19. Coupling the fixation location of the eyes on the
positioning planes to the presented stimulus.
20. Correction of the calculated fixation locations, of
the eyes on the presented stimuli, for the measured and
calculated 3-dimensional orientation, position and situation
of the stimulus.
CA 02248672 1998-09-10
WO 97/38624 PCT/NL97/00188
17
All steps are performed such, that a net time resoluti-
on of at least 1/50 of a second and a location resolution
(on the stimuli) of at least 1 mm2 are realized.
By way of an example, it is indicated step by step
which aspects, among others, can be measured with the system
and method according to the invention. The aspects concern
items of typical printed matter research.
1. A diagnosis of respondents, e.g. itemized in:
- sex,
- age,
- education,
- demographic variables,
- social variables,
- areas of interest,
- preferences, e.g. in relation to brands and products,
- which magazines are read,
- habits, e.g. smoking and drinking,
- visual, auditive and psychological capacities,
- absorbing capacity,
- reading capacity,
- understanding capacity,
- tempo,
- calibration, etc.
2. A diagnosis of stimuli, e.g. print media, itemized in:
- number and type of magazines,
- number of pages,
- number of pages opened and the page numbers,
- number of pages respondents left opened for more than
X seconds, and page numbers,
- pages on which respondents placed more than X fixat-
ions, and page numbers,
- number of fixations per page, per quadrant page, or
smaller, e.g. 1/9 of a page.
- time spent for a complete magazine, per page, per
quadrant page, or smaller, e.g. 1/9 of a page, etc.
- Above diagnoses itemized seperately for:
- advertisement pages,
- editorial pages,
- left pages,
CA 02248672 1998-09-10
WO 97/38624 PCT/NL97/00188
18
- right pages,
- front and back pages, etc.
3. The aspects of stimuli, such as:
- orientation in medium: date, issue, page number,
location,
- subject, theme,
- size,
- style, colour use, etc.
Performance aspects:
- The degree to which stimuli, e.g. advertisements, are
capable of realizing an observation by respondents,
- To what extent all parts of an advertisement regarded
as necessary by the advertiser are fixed by respond-
ents,
- The number of persons that fixated one, two, or three
of the three central advertisement elements,
- The degree to which stimuli, e.g. advertisements, are
(in)capable of establishing contact with respondents.
- Ad contact,
e.g. defined as: the percentage of respondents
who placed at least one or more fixations on the
ad according to certain time criteria.
Ad contact means the start of taking in the
stimulus.
- No ad contact,
e.g. defined as: the percentage of respondents
who did not open the double page, on which the
ad was placed, or did not place any fixations on
the advertisement according to certain time
criteria.
The performance aspects following below are
related to whether or not fixating one or more of
the three central advertisement elements.
The three central advertisement elements (C.A.E.)
are:
- Brand in parts
- Visual in parts
- Text in parts
CA 02248672 1998-09-10
WO 97/38624 PCT/NL97/00188
19
- Partial Advertisement Contact (PAC)
e.g. defined as: the percentage of respondents
having fixated the brand + the two other central
advertisement elements, each with individual time
criteria, e.g. between 60 and 1000 ms.
Partial Advertisement Contact is a requirement
for the capability to take note of a part of the
brand message, in such a way that absorption
thereof becomes possible.
- Complete Advertisement Contact (CAC)
e.g. defined as: the percentage of respondents
having fixated the brand + the two other central
advertisement elements, each with individual time
criteria, e.g. between 60 and 1000 ms.
Complete Advertisement Contact is a necessary
requirement for the capability to fully absorb,
process and store the advertisement information.
- The Advertisement Contact Score (ACS)
e.g. defined as: the result of Partial and Complete
together.
The Advertisement Contact Score percentage indica-
tes the number of respondents who could have ab-
sorbed the stimulus in such a way, that at least a
correct, be it a partial one, recall is possible.
The criteria to be employed, e.g. the ACS criteria,
are preferably adjustable. Thus, a difference in
requirements can be employed e.g. for known stimuli
in relation to unknown stimuli.
All data are relative in relation to:
- other stimuli.
- the same stimulus in other media.
-'the same stimulus at other times.
4. The circumstances during the measurement, such as:
- date and moment of the day.
- the moment of measurement,
e.g. separation of morning, afternoon and evening
results.
- the total number of respondents,
- the total number of magazines and/or papers,
CA 02248672 1998-09-10
WO 97/38624 PCT/NL97/00188
- the total number of pages,
- the total number of advertisements,
- the news situation (including topical matters and
sports), etc.
5 - the social-economical situation,
- the temperature (indoors and outdoors),
- the atmospheric humidity level and atmospheric
pressure,
5. Any conceivable combination of 1, 2, 3 and 4.
10 It will be obvious that in the above, the invention has
only been explained by way of some specific examples, as for
the method, the equipment used with it, the applications and
the possible result of measurements, and that many changes
and/or additions can be made without leaving the inventive
15 idea.
- claims -
