US20210022637A1 - Method for predicting efficacy of a stimulus by measuring physiological response to stimuli - Google Patents

Abstract

Methods are disclosed for determining an efficacy of a stimulus based on one or more measurable physiological responses to one or more stimuli including one or more stimulus features. Data is acquired on physiological responses of a group of one or more subjects to presentation of one or more stimuli including one or more stimulus features. The data on the one or more physiological responses of the one or more subjects is correlated with the presentation of the one or more stimulus features included in the one or more stimuli. The correlated data on the one or more physiological responses are associated with a separately-determined efficacy of the one or more stimuli to form a stimulus efficacy model. From this information, a projected efficacy of a stimulus is determinable by comparing one or more subsequently-measured physiological responses to the stimulus with the stimulus efficacy model.

Description

RELATED APPLICATION
• This application claims the benefit of U.S. Provisional Patent Application Ser. No. 62/041,706, titled “METHOD FOR MEASURING PHYSIOLOGICAL IMPACT OF STIMULUS FEATURES TO PREDICT RESPONSE OF SUBJECTS TO STIMULI INCLUDING SUCH STIMULUS FEATURES,” filed on Aug. 26, 2014, the entire contents of which are hereby incorporated in its entirety herein.
BACKGROUND
• Every day, countless sums of money are expended in trying to influence opinions or behavior of individuals or groups, or to determine what types of images, sounds, or other stimuli may successfully influence the opinions or behavior of individuals or groups. Just to list a few examples, movie producers, advertisers, politicians, video game developers, and many others invest huge amounts of time and money in trying to engage their audience and generate a positive response to their offerings. To this end, for instance, a great deal of time, money, and energy is invested in conducting focus groups, administering consumer surveys, evaluating user behavior, and in conducting other processes in the attempt to create media events that will attract consumers or other supporters.
• Despite the desire to be able to predict a response to media stimuli or other stimuli, it has proven difficult to make predictions of what stimuli will engage an audience and elicit a desirable response. Existing methods often prove unsatisfactory. For example, in focus groups, participants may be asked to rate or rank advertisements based on their individual preferences and the quality of the ads. However, asking participants to rate or rank advertisements may not yield accurate responses for several reasons. First, when participants are brought into a focus group or otherwise asked to watch the advertisements, they are actively requested to watch the advertisements and, thus, may react differently than ordinary viewers, who may passively watch the advertisement when it is inserted in a break within a program. The different mindset can result in the focus group having a very different response to the advertisement than an ordinary viewer. Similarly, being asked to deliberately form an opinion or a preference may bias a subject to report what the subject believes that the person administering wants to hear.
• It thus would be a significant advance in the art to be able to derive a more accurate way of gauging a response of one or more subjects to a set of stimuli as may be included in an audio and/or video an advertisement, a television program, video game, a movie, music, or other media, or in any other form of stimuli.
SUMMARY
• Methods are disclosed for determining an efficacy of a stimulus based on one or more measurable physiological responses to one or more stimuli including one or more stimulus features. Data is acquired on physiological responses of a group of one or more subjects to presentation of one or more stimuli including one or more stimulus features. The data on the one or more physiological responses of the one or more subjects is correlated with the presentation of the one or more stimulus features included in the one or more stimuli. The correlated data on the one or more physiological responses are associated with a separately-determined efficacy of the one or more stimuli to form a stimulus efficacy model. From this information, a projected efficacy of a stimulus is determinable by comparing one or more subsequently-measured physiological responses to the stimulus with the stimulus efficacy model.
• In another aspect, a method to project an efficacy of a stimulus based on one or more measurable physiological responses is disclosed. Data is acquired on physiological responses of a group of one or more subjects to presentation of one or more stimuli included one or more stimulus features. A first data matrix is formed, where the first data matrix includes one or more stimuli tracked against a time over which the one or more stimulus features included in the one or more stimuli were presented. A second data matrix including one or more physiological responses of the one or more subjects to which the one or more stimulus features included in the one or more stimuli were presented tracked against the time over which the one or more stimulus features included in the one or more stimuli were presented. A separately-determined efficacy of the one or more stimuli as determined from at least one of the one or more subjects and an additional group of one or more subjects is associated with the one or more resulting physiological responses. Statistical processing is applied to generate a model associating the separately-determined efficacy of the one or more stimuli with the one or more physiological responses to determine a projected efficacy of a stimulus based on subsequently-measured physiological responses to the stimulus.
• In still another aspect, a method of predicting a response of one or more subjects to a stimulus includes subjecting a group of one or more subjects to one or more stimuli including one or more stimulus features. Data is acquired on one or more physiological responses of the group of one of more subjects to presentation of the one or more stimulus features included in the one or more stimuli. An efficacy of the stimuli is determined from at least one of the one or more subjects and an additional group of one or more subjects. A model is generated correlating the one or more physiological responses to the presentation of the one or more stimulus features included in the one or more stimuli. The model is associated with the efficacy of the one or more stimuli to form a stimulus efficacy model. The stimulus efficacy model may be used to predict efficacy of a stimulus by measuring one or more physiological responses of prospective subjects to the stimulus and comparing the one or more physiological responses of the prospective subjects with the stimulus efficacy model.
• Other methods are further described in detail below.
BRIEF DESCRIPTION OF THE DRAWINGS
• FIGS. 1 and 2 are schematic diagrams depicting analysis of responses of a group of one or more subjects to one or more stimuli including one or more stimulus features according to aspects of the disclosure;
• FIGS. 3 and 4 are block diagrams of systems used to analyze responses of one or more subjects to one or more stimulus features included in one or more stimuli according to aspects of the disclosure;
• FIG. 5 is a schematic diagram of a process of assessing separately-determined efficacy of one or more stimuli according to aspects of the disclosure;
• FIGS. 6-8 are flow diagrams of exemplary methods of acquiring and/or using data on physiological responses of one or more subjects to one or more stimuli including one or more stimulus features according to aspects of the disclosure; and
• FIG. 9 is a block diagram of an exemplary computing system that may be used to acquire and process data on physiological responses to one or more stimuli including one or more stimulus features and data on separately-determined efficacy of the one or more stimuli according to aspects of the disclosure.
DETAILED DESCRIPTION
• Methods of the present disclosure assist in predicting the response of one or more subjects to one or more visual or audible stimuli including one or more stimulus features to be presented to the one or more subjects.
• FIG. 1 is a schematic diagram of aprocess100 depicting an analysis of physiological responses of a group of one ormore subjects110 topresentation120 of one or stimuli121. In the process to acquire data on physiological responses to one ormore stimuli124, the group of one ormore subjects110 is subjected to one or more stimuli121, such as visual or auditory stimuli. The one or more stimuli121 possess one or more stimulus features122. For example, an auditory stimulus, such as a particular sound, may include stimulus features such as pitch trajectory, sound envelope, speech envelope, spectral power, rhythm, volume, and pitch. It is known that a human mind and body have measureable reactions to stimuli, including changes in heart rate, respiration rate, eye activity, brain activity, and other biophysical reactions. A process to measure and record data on thephysiological responses123 may include electroencephalography (EEG), electrocardiography (EKG), galvanic skin response (GSR), and other techniques to measure heart rate, respiration rate, eye activity, or other physiological responses to stimuli.
• A process130 correlates the data acquired on the physiological responses with the presentation of the one or more stimulus features included in the one or more stimuli, such as by correlating a time at which the one or more stimulus features122 of the one or more stimuli121 were presented with the acquired physiological responses. Using statistical processes described further below with reference toFIGS. 3 and 4, aprocess140 is used to associate the correlation between the resulting one or more physiological responses and the one or more stimuli121 including the one or more stimulus features122 to a separately-determined efficacy of the one ormore stimuli145. The separately-determined efficacy of the one ormore stimuli145 is ascertained from the group of one ormore subjects110 and/or a separate group of one or more subjects. The separately-determined efficacy data is determined by polling subjects, interviewing subjects, subjecting subjects to testing to determine their responses while being subjected to one or more stimuli, as further described with reference toFIG. 5.
• The determined association of the degree of correlation between the physiological responses and the one or more stimuli121 including the one or more stimulus features122 with the separately-determined efficacy of the one or more stimuli121 may be used to create astimulus efficacy model150. Thestimulus efficacy model150 thus may be used in a process to determine a projected efficacy of astimulus160. In such aprocess160, subsequently-acquired physiological data acquired in response to a stimulus may be used to determine the efficacy of that stimulus to determine whether the stimulus is desirable for inclusion in an audio, visual, or audiovisual presentation, such as an advertisement, a program, a video game, etc. Thus, using thestimulus efficacy model150 generated via theprocess100 may enable one to create a media presentation that includes a stimulus or stimuli that have a desired efficacy.
• FIG. 2 is a schematic diagram of aprocess200 depicting an analysis of responses of a group one ormore subjects210 topresentation220 of one ormore stimuli221 including one or more stimulus features222 according to another aspect of the disclosure. In theprocess200, while the group of one ormore subjects210 is presented220 with the one ormore stimuli221 including the one or more stimulus features222,physiological responses230 of the group of one ormore subjects210 are acquired as described with reference toFIG. 1. The magnitude of the physiological responses may be included in the acquiredphysiological data230. In addition, a separately-determined efficacy245 of each of the one ormore stimuli221 is determined. As previously stated, a process by which a separately-determined efficacy of the one ormore stimuli221 is described with reference toFIG. 5.
• In order to develop a model that may identify physiological responses to the one or more stimulus features222 included in the one ormore stimuli221, afirst data matrix240 tracking the presentation of the one ormore stimuli221 including the one or more stimulus features222 is created, tracking the one ormore stimuli221 and the one or more stimulus features222 over a period of time during which they are presented. Also, asecond data matrix250 tracking the physiological responses of the group of one ormore subjects210 is formed, tracking the measured responses over the time period during which the one or more stimulus features222 included in the one ormore stimuli221 are presented. The collected data in thefirst matrix240 may include aquantified magnitude235 of each of the one or more stimulus features222 included in the one ormore stimuli220, and thesecond matrix250 may incorporate the quantified degree of the physiological responses to the one or more stimulus features222 included in the one ormore stimuli221 in the acquiredphysiological data230.
• Thefirst data matrix240 and thesecond data matrix250 are subject tocorrelation processing260, which may include statistical regression of the physiological responses to the one or more stimulus features222 included in the one ormore stimuli221. The correlation processing results in amodel270 that includes the physiological responses correlated with the stimuli.
• As described with reference toFIG. 1, the model can be used to evaluate whether any of the stimuli121 is likely to achieve a desired effect by measuring the physiological responses of subjects to the one or more stimuli121. The resulting model may be consulted to determine what type or stimuli should be included in an audio, visual, or audiovisual work being developed so that the work will have a desired effect on its audience.
• FIG. 3 is a block diagram of asystem300 used to analyze responses of one or more subjects to sets of stimuli including one or more stimulus features according to an aspect of the disclosure. One or more subjects, such assubject310, are subjected to astimulus320, which may incorporate one or more identifiable stimulus features. Thestimulus320 may, for example, include an audio, visual, or audiovisual presentation that includes one or more stimulus features. Afeature extraction subsystem325 identifies a plurality of features, such as features f₁326,f₂327, andf_n328, and captures each as a function of time. Data representing the extracted features are combined by anaggregator345. As further described below, afilter435 applies weights, such as weights w₁336,w₂337, andw_n338, are assigned to the extracted feature data to optimize the yield of the model. The resulting output is weighted, extracted feature data tracked over time, y₁(t)357.
• In a particular aspect, the features are used as independent variables to which resulting physiological responses are correlated as dependent variables. In a visual presentation or a visual component of an audiovisual presentation, features extracted may include, for example, orientation energy, optical flow, luminance and chrominance, local spatial contrast, and temporal contrast. For an audio presentation or an audio component of an audio visual presentation, the features extracted may include pitch trajectory, sound envelope, speech envelope and spectral power or powers. The physiological responses may include data yielded by electroencephalography (EEG), electrocardiography (EKG), galvanic skin response, heart rate monitoring, respiration rate monitoring, eye tracking, and pupillometry.
• Correspondingly, the subject310 is presented with thestimulus320 while being monitored by physiological measuring devices. InFIG. 3, the subject310 is being monitored by an electroencephalography system that monitors physiological responses of the subject over time, including spatially-arrayed input leads features x₁(t)311, x₂(t)312, and x_n(t)313. Afilter315 applies weights, such as weights v₁316,v₂317, andv_n318 to the measured responses. Again, as further described below, the weights are assigned to the extracted feature data to optimize the yield of the model. Data representing the physiological responses are combined by anaggregator325. The resulting output is weighted, physiological response data tracked over time, y₂(t)359.
• The weighted, extracted feature data tracked over time, y₁(t)357, and the weighted, physiological response data tracked over time, y₂(t)359, is submitted to acorrelator355 and then anoptimizer365. Output of theoptimizer365 is fed back to thefilters315 and335 to adjust the applied weights to improve the signal-to-noise ratio inherent in the correlated data, as well as to account for delays resulting from a subject's physiological response to various stimuli. In a particular aspect, a block-Toeplitz structure is used so that a temporal filter represented by the weights applied by thefilter335 filters across a temporal aperture, allowing for short-time integration of the multiple stimuli. In a particular aspect, a spatiotemporal matrix of EEG responses is constructed such that thefilter315 linearly filters one or more underlying neuronal sources. Temporally filtering the stimuli while spatially filtering the neuronal responses leads to pairs of optimal filters which uncover the covariation of specific temporal frequencies of the stimulus to specific neuronal sources or patterns of sources in the brain. For example, activity in an occipital component of the EEG may be found to covary with a visual stimulus feature, while activity in a temporal component may be found to covary with an audio stimulus feature. Each of these pairs is associated with an optimal correlation value, such as a generalized eigenvalue that conveys a level of physiological drive elicited by a selected identified stimulus.
• The system may employ a variety of physiological responses recorded before, during and after a given stimulus. This includes, but is not limited to, recordings from EEG, EKG, Eye Tracking, Pupillometry, Heart Rate/Heart Rate Variability, GSR, and Respiratory Rate. A variety of models can be used to relate the physiological responses to the stimuli, including, but not limited to, temporal filters, spatial filters, spectral models, neural networks, and deep (learning) networks. These models all serve to learn an optimal representation between a set of computed stimuli and a corresponding set of physiological signals.
• FIG. 4 is a block diagram of asystem400 that may be used to derive inter-subject correlation among a plurality of subjects. Thesystem400 shares many common elements with thesystem300 ofFIG. 3. However, thesystem400 focuses on correlating physiological responses betweensubjects405 and410 in response to a stimulus including one or more stimulus features401, rather than correlating physiological responses of a subject to one or more stimuli, as is the case inFIG. 3. It should be appreciated that thesystems300 and400 may be combined to inter-relate responses of multiple subjects to a plurality of stimuli.
• Each of the one ormore subjects405 and410 are subjected to thestimulus401. Thesubjects405 and410 are monitored by electroencephalography systems that monitor physiological responses of the subject over time. The subject405 is monitored by spatially-arrayed input leads features x₁₁(t)406, x₁₂(t)407, and x_1n(t)408. Afilter435 applies weights, such as weights v₁₁436,v₁₂437, andv_1n438 to the measured responses. Data representing the physiological responses are combined by anaggregator445. The resulting output is weighted, physiological response data tracked over time, y₁(t)457. The subject410 is monitored by spatially-arrayed input leads features x₂₁(t)411, x₂₂(t)412, and x_2n(t)413. Afilter415 applies weights, such as weights v₂₁416,v₂₂417, andv_2n418 to the measured responses. Data representing the physiological responses are combined by anaggregator425. The resulting output is weighted, physiological response data tracked over time, y₂(t)459.
• The weighted physiological response feature data tracked over time, y₁(t)457 and the weighted, physiological response data tracked over time, y₂(t)459 are submitted to acorrelator455 and then anoptimizer465. Output of theoptimizer465 is fed back to thefilters415 and435 to adjust the applied weights to improve the signal-to-noise ratio inherent in the correlated data, as well as to account for delays resulting from a subject's physiological response to various stimuli. The inter-subject data is correlated with the stimulus feature data (not shown inFIG. 4) to yield physiological response data collected from multiple subjects to the stimulus feature data.
• FIG. 5 is a schematic diagram of aprocess500 of assessing separately-determined efficacy of one or more stimuli. To provide a body of information for association with the correlation between the physiological data and the stimuli, a group of one ormore subjects510, which may be the same group from which physiological data is acquired as described with reference toFIGS. 1 and 2, and/or a second group of one ormore subjects511, are subjected to aprocess520 in which one ormore stimuli522 are presented. The stimuli, as previously described, may include one or more visual stimuli, audible stimuli, or a combination of visual and audible stimuli. Upon being subjected to the one or more stimuli subjects in the first group and/or the second group (if more than one group is used) are subjected to aprocess530 to evaluate their response to the one or more stimuli. For example, the subjects may be polled regarding their subjective responses to the one or more stimuli. Also, the one or more subjects may be interviewed individually or as part of a group (e.g., a focus group) regarding their responses to the one or more stimuli. Further, the subjects may be observed to acquire their reactions to the one or more stimuli as the stimuli are presented. Alternatively, the one or more subjects may be submitted to testing to gauge their performance subsequent to or during the presentation of the stimuli to determine whether the stimuli have an effect on the subjects' cognitive performance on such testing. Additional methods of evaluating the response of subjects to the one or more stimuli also may be used. It should be appreciated that one or more of these methods of evaluating the subjects' responses may be used.
• FIG. 6 is a flow diagram of anexemplary method600 of analyzing responses of one or more subjects to sets of stimuli according to aspects of the disclosure. At602, data is acquired on one or more physiological responses of one or more subjects to presentation of one or more stimuli including one or more stimulus features. As previously described with reference toFIGS. 1-4, the data may be acquired by EKG, GSR, heart rate, respiration rate, eye-gaze, fixation duration, pupilometry, or other measurements. At604, the data on the one or more physiological responses of the one or more subjects are correlated with the presentation of the one or more stimulus features included in the one or more stimuli. At606, the correlated data on the one or more physiological responses are associated with a separately-determined efficacy of the one or more stimuli to form a stimulus efficacy model. As a result, a projected efficacy of a stimulus is determinable from one or more subsequently-measured physiological responses to the stimulus.
• FIG. 7 is a flow diagram of anexemplary method700 of analyzing responses of one or more subjects to sets of stimuli according to aspects of the disclosure. At702, data is acquired on physiological responses of one or more subjects to presentation of one or more stimuli including one or more stimulus features. As previously described with reference toFIGS. 1-4, the data may be acquired by EKG, GSR, heart rate, respiration rate, eye-gaze, fixation duration, pupilometry, or other measurements. At704, a first data matrix is formed, where the first matrix includes one or more stimulus features included in the one or more stimuli tracked against a time over which the one or more stimuli were presented. At706, a second data matrix is formed that includes one or more physiological responses of a first group of one or more subjects that were exposed to the one or more stimulus features included in the one or more stimuli tracked against the time over which the one or more stimuli were presented. At708, a separately-determined efficacy of the set of stimuli as determined from at least one of the first group of one or more subjects and a second group of one or more subjects are associated with the one or more resulting physiological responses. The separately-determined efficacy is measured by polling, interviewing, observing, testing, or other methods as described with reference toFIG. 6. At710, statistical processing is applied to correlate the separately-determined efficacy of the set of stimuli with the one or more physiological responses to determine a projected efficacy of a stimulus based on subsequently-measured physiological responses.
• FIG. 8 is a flow diagram of another exemplary method of analyzing responses of one or more subjects to sets of stimuli according to aspects of the disclosure. At802, one or more subjects are subjected to one or more stimuli including one or more stimulus features. At804, data on one or more physiological responses of the one of more subjects to the one or more stimulus features included in the one or more stimuli are acquired. As previously described with reference toFIGS. 1-4, the data may be acquired by EKG, GSR, heart rate, respiration rate, eye-gaze, fixation duration, pupilometry, or other measurements. At806, efficacy of one or more stimuli from at least one of the one or more subjects and an additional group of one or more subjects is determined. The separately-determined efficacy is measured by polling, interviewing, observing, testing, or other methods as described with reference toFIG. 5. At808, a model correlating the one or more measurable physiological responses to the presentation of the one or more stimulus features included in the one or more stimuli is generated. At810, the model is correlated with the efficacy of the one or more stimuli. The model may be used to determine prospective efficacy of a stimulus by subjecting one or more additional subjects to the stimulus and measuring the one or more measurable physiological responses of the one or more additional subjects.
• FIG. 9 is a block diagram of anexemplary computing system900 that may be used for performing computer-implemented methods or executing computer-executable instructions for generating or using a model correlating physiological responses with one or more stimuli including one or more stimulus features according to the present disclosure.
• Referring toFIG. 9, thecomputing system900 may include any of a number of forms of stationary or mobile computing devices. Thecomputing device900 typically includes at least oneprocessing unit904 and asystem memory906. Depending on the exact configuration and type of computing device, thesystem memory906 may be volatile (such as RAM), non-volatile (such as ROM, flash memory, and the like) or some combination of the two. Thesystem memory906 typically maintains anoperating system908, one ormore applications910, andprogram data912.
• Thecomputing device900 may also have additional features or functionality. For example, thecomputing device900 may also include additional data storage devices (removable and/or non-removable) such as, for example, magnetic disks, optical disks, or tape. Such additional storage is illustrated inFIG. 9 byremovable storage914 andnon-removable storage916. Computer storage media may include volatile and non-volatile, removable and non-removable media implemented in any method or technology for storage of information, such as computer readable instructions, data structures, program modules or other data. Thesystem memory906, theremovable storage914, and thenon-removable storage916 are all examples of computer storage media. Available types of computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory (in both removable and non-removable forms) or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to store the desired information and which can be accessed by thecomputing device900. Any such computer storage media may be part of the computing device1000.
• The computing device1000 may also have input device(s)918 such as a keyboard, mouse, pen, voice input device, touch input device, etc. Further, in accordance with aspects of the present disclosure, physiological measurement devices to track EEG, EKG, heart rate, respiration rate, eye-gaze, fixation duration, pupilometry, or other measurements may be operatively coupled as input devices to thecomputing device900 to acquire physiological data. Output device(s)920 such as a display, speakers, printer, etc., may also be included. In accordance with aspects of the disclosure, such devices may be used to present the one or more stimuli. Thecomputing device900 thus may be used to present stimuli, capture physiological data, and correlate the data with the stimuli and included stimulus features. Thecomputing device900 may also be used to receive separately-determined data on the efficacy of the stimulus. Similarly, suitable computer-readable instructions may be used to generate models and perform statistical processing on the data in accordance with aspects of the disclosure.
• Thecomputing device900 also may include one ormore communication connections922 that allow thecomputing device900 to communicate withother computing devices924, such as over a network or a wireless network. The one ormore communication connections922 are an example of communication media. Available forms of communication media typically carry computer readable instructions, data structures, program modules or other data in a modulated data signal such as a carrier wave or other transport mechanism and includes any information delivery media. The term “modulated data signal” may include a signal that has one or more of its characteristics set or changed in such a manner as to encode information in the signal. By way of example, and not limitation, communication media may include wired media such as a wired network or direct-wired connection, and wireless media such as acoustic, RF, infrared and other wireless media. The term computer readable media as used herein includes both storage media and communication media.
• While the disclosure has been has been set forth herein in reference to specific aspects, features and illustrative aspects, it will be appreciated that the utility of the disclosure is not thus limited, but rather extends to and encompasses numerous other variations, modifications and alternative aspects, as will suggest themselves to those of ordinary skill in the field of the present disclosure, based on the description herein. Any of various elements or features recited herein is contemplated for use with other features or elements disclosed herein, unless specified to the contrary. Correspondingly, the invention that may be hereinafter claimed is intended to be broadly construed and interpreted, as including all such variations, modifications and alternative aspects, within its spirit and scope.

Claims (10)

1.-20. (canceled)

21. A method to predict an efficacy of a stimulus based on one or more previously-measured physiological responses to one or more stimuli including one or more stimulus features, the method comprising:

acquiring data on physiological responses of one or more subjects to presentation of one or more stimuli including one or more stimulus features;

forming a first data matrix tracking the one or more stimulus features included in the one or more stimuli tracked against a time over which the one or more stimulus features included in the one or more stimuli were presented;

forming a second data matrix including one or more physiological responses of the one or more subjects to which the one or more stimulus features included in the one or more stimuli were presented tracked against the time over which the one or more stimulus features included in the one or more stimuli were presented;

associating a separately-determined efficacy of the one or more stimuli as determined from at least one of the one or more subjects and an additional group of one or more subjects with the one or more resulting physiological responses; and

applying statistical processing to correlate the separately-determined efficacy of the one or more stimuli with the data on the one or more physiological responses to determine a projected efficacy of a stimulus based on subsequently-measured physiological responses to the stimulus.

22. The method ofclaim 21, wherein the data on the one or more physiological responses are acquired by one or more of:

electroencephalography (EEG);

electrocardiography (EKG);

galvanic skin response;

heart rate monitoring;

respiration rate monitoring;

eye tracking; and

pupillometry.

23. The method ofclaim 22, wherein when the data on the one or more physiological responses are measured by electroencephalography, further comprising forming the model by temporally filtering the presentation of each of the one or more stimulus features included in the one or more stimuli by spatially filtering the one or more physiological responses from the electroencephalography.

24. The method ofclaim 21, wherein a statistical process is used to regress the first data matrix of onto the second data matrix.

25. The method ofclaim 24, wherein the statistical process includes canonical correlation analysis.

26. The method ofclaim 26, wherein correlation between the one or more stimulus features included in the one or more stimuli and the one or more physiological responses is represented by one or more eigenvalues.

27. The method ofclaim 21, wherein correlating the data on the one or more physiological responses of the one or more subjects includes inter-subject correlation between the one or more subjects.

28. The method ofclaim 27, wherein the inter-subject correlation includes a group-level inter-subject correlation.

29. (canceled)

Images