Patent Application
20080222670
1. Field of Invention
This invention relates to the field of media performance rating based on physiological responses from viewers.
2. Background of the Invention
Prior methods and systems for calculating how viewers respond to a media often include taking an average response from a survey and/or adopting viewer “knobs” or other rating schemes. These prior methods show only what the average response is, not how individuals respond. Just as individuals' responses can be strongly biased by prior experienced personal events, their responses to an event in the media can also be influenced by a previous event that happened directly before the current event. This effect can heavily skew the average response to the current event in a manner not directly related to the current event. Consequently, by just taking the average of physiological responses across individual viewers into account at a given time, the prior approaches may be biased by subjective responses of viewers and are incapable of capturing the whole picture of how viewers feel when they view the media, as individual outlier responses dominate the average response.
Various embodiments of the present invention create a novel system and method for rating an event in a media based on the strength of the emotions viewers feel towards the event. The viewers' responses to the media can be measured and calculated via physiological sensors. The metric for rating the strength of the media is created based on the mathematical coherence of change (up or down) of all pertinent physiological responses across the viewers. Such rating offers an objective ability to compare the strengths of events of the media, as there is a strong correlation between high coherence of physiological responses (all viewers feel the same thing at the same time) and strong ratings of emotionality, engagement, likeability, and success in the marketplace/on screen.
a)-(c) show an exemplary integrated headset used with one embodiment of the present invention from different angles.
a)-(b) show exemplary coherence of response (CoR) values for advertisements and video games, respectively, in accordance with one embodiment of the present invention.
The invention is illustrated by way of example and not by way of limitation in the figures of the accompanying drawings in which like references indicate similar elements. It should be noted that references to “an” or “one” or “some” embodiment(s) in this disclosure are not necessarily to the same embodiment, and such references mean at least one.
Various embodiments of the present invention create a method and system to rate the performance of a media using the coherence of physiological responses across groups of viewers of the media. The coherence of response metric objectively measures the strength of the media across many people as it shows when viewers feel the same emotions at the same time. Compared to the survey data or average value of physiological responses, the coherence of the physiological responses corresponds to viewers either having the same emotional responses at the same time, or having different ones at the same time, which is a much better indicator of performance of the media.
The rating approach of the present invention is based on the principle that, in the marketplace, a media that performs strongly is not only capable of creating strong emotional responses in one viewer, but is also capable of creating the same strong emotional responses across many viewers. Viewers feel different emotions while they watch a piece of media, but for a successful media, there are key events in the media that create a coherent response in physiological data across groups of viewers. A media will not perform well if there is no coherent response to the media; i.e., the media will not sell well because of bad ratings, bad word of mouth or other negative views if some viewers really like the media while other comparable viewers do not.
Referring to
Once measured, the physiological data of the viewers can be transmitted to a coherence module 104, which derives physiological responses of the viewers based on their physiological data measured while they were watching the event. Here, the physiological response can be one or more of: thought, liking, engagement, immersion, physical engagement, valence, and vigor, wherein thought and liking can be calculated from EEG. Coherence of the viewers' physiological responses to the event can then be calculated and the event can be rated based on the coherence of the physiological responses from the viewers. Optionally, the media itself can also be rated based on the ratings of all the events in the media.
Referring to
In some embodiments, an integrated headset can be placed on a viewer's head for measurement of his/her physiological data while the viewer is watching an event of the media. The data can be recorded in a program on a computer that allows viewers to interact with media while wearing the headset.
In some embodiments, the integrated headset can be turned on with a push button and the viewer's physiological data is measured and recorded instantly. The data transmission can be handled wirelessly through a computer interface that the headset links to. No skin preparation or gels are needed on the viewer to obtain an accurate measurement, and the headset can be removed from the viewer easily and can be instantly used by another viewer. No degradation of the headset occurs during use and the headset can be reused thousands of times.
In some embodiments, the viewers' physiological responses can be derived via a plurality of formulas, which use the physiological data of the viewers as inputs. Facial expression recognition, “knob” and other measures of emotion can also be used as inputs with comparable validity. Each of the derived physiological responses, which can include but are not limited to, “Engagement,” “Adrenaline,” “Thought,” and “Valence,” combines physiological data from multiple sensors into a multi-dimensional, simple-to-understand, representation of viewers' emotional response.
In some embodiments, the physiological responses can be calculated across many different viewers as they are watching the same media, either at the same time or at different times. The physiological responses can be treated as a time series for each viewer, and can be combined for multiple viewers to show aggregate responses. The physiological responses can also be averaged or viewed over sections of the data to analyze the responses to key events in the media. As a non-limiting example,
In some embodiments, the physiological responses from viewers can be analyzed specific to the type of media the viewers are watching. For non-limiting examples:
In some embodiments, the coherence module is operable to calculate Coherence of Response (CoR) of the viewers based on time series of physiological data recorded while the viewers are interacting with the media. Coherence of response can be calculated for all points in the time series based on how closely the trends in the responses are to each other at every given point. There are many ways to mathematically calculate the trend in the responses in a time series, which include, for non-limiting examples, the standard derivation approach and the binning approach.
In some embodiments, the coherence of the physiological responses can be calculated based on standard deviation of the physiological responses and/or standard deviation of a derivative of the physiological responses (e.g., the magnitude and value of the physiological responses) from the plurality of viewers during a specific one event. A correlation can be calculated against the derivative of all the physiological responses and the change in a trend is reflected by the number of viewers with a positive derivative of an individual response at a given time or duration of time of the event.
In some embodiments, physiological responses from the viewers to either the entire media or an event in the media can be categorized and binned into predetermined categories based on similarity of the responses. Here, the predetermined categories can include one or more of: very positive, positive, flat, negative, very negative. For a non-limiting example, responses from viewers with positive change across the event are categorized in one bin and responses from viewers with non-positive change across the event are categorized in another bin. The number/percentage of viewers that respond in one way or the other are then counted. For instance, the number of viewers responding with positive emotions to a joke (during most time of the joke) are counted and compared to the number of viewers responding without positive emotions to the same joke. Other embodiments utilize different bins or different response data, including small/large deviation where small and large are defined by being below and above a predetermined number.
Once the physiological responses from the viewers are binned, the overall coherence of the responses to the event and/or the media can be rated based on the distribution of number of viewers in each response bin. In some embodiments, the percentage of viewers responding with a positive response change is used to rate an event. For a non-limiting example, if 80% of viewers respond positively to a joke, the joke is considered as successful. Other embodiments aggregate the ratings across many events (e.g. the percentage of events with over 70% coherence in change) and other metrics.
As a non-limiting example,
In some embodiments, time series of physiological data can also be smoothed and filtered prior to CoR calculations to allow for different scales of responses to be calculated. For a non-limiting example, if the physiological data is recorded at 100 Hz, but a calculation for a long event (e.g., 2 minutes) is needed, the high frequency data can be removed, allowing the information in the data to be of the same order of magnitude in frequencies as the length of calculation. In addition, if CoR needs to be calculated over a short period of time (e.g., a few seconds), the low frequency components of the signals can be removed to minimize the influence of prior events that could affect the position of the data. This approach will line up the data between viewers and allow only the pertinent data to be used in the CoR calculation.
Rating Media with CoR
Once CoR are calculated at every point in the time series of physiological responses for a set of physiological responses to an event, a rating can be created for the event and/or the media via one or more of the following:
In some embodiments, scores can also be calculated for non-linear, or semi-linear media such as video games, websites and other interactive media in addition to scores calculated based on a time series that is constant between viewers. The time series for each viewer can be divided into events that appear across all viewers. Such events include but are not limited to fight scenes in video games, explosions, viewing the home page on a website, going to a chapter of a DVD and any other ones.
In some embodiments, the coherence module can create an overall rating for CoR by averaging the magnitude of the CoR values of various events in the entire media in order to compare the strength of different events in the media:
Non-linear weights can also be placed on the CoR value:
Here, f(x) can be a function such as log, polynomials or other ones:
f(x)=x2
f(x)=x3
f(x)=x2+x
f(x)=log(x)
Here, the function can change the characteristics of the data, which greatly increases the differentiation seen in the score. This pertains to low coherence scores, such as 10% coherence being classified as very low, and high coherence scores such as 70% coherent being classified as very good. By using a function such as f(x)=x2, these two values will be very strongly differentiated, making the 10% value only 0.01 and the 70% value 0.49, or 5 times bigger. This differentiates high coherence points from low coherence points, penalizing a media more for being low coherence and helping a media for being high coherence. Overall, CoR score is a value that can be used to compare the experience and strength of multiple events in the media and other metrics of this kind. The higher this value is, the more coherent the media is.
The coherence of responses to each instance of an event can then be calculated via the same mathematical formulas, with the period of calculation being the length of the event instead of increments of time. The outcome of this will be a rating of responses to each event. For a non-limiting example, if 70% of events where a video game player for game X interacted with boss monster X lead to a rise in engagement, while 90% of interactions with boss monster Y lead to rises in engagement, boss monster Y has a more coherent response than boss monster X. Such calculation can be done for every pertinent event in the piece of media and these scores can then be combined to create overall scores.
In some embodiments, the coherence module can create a more complete measure of the coherence of the experience by combining the CoR values for multiple vectors of physiological responses:
Full Emotional CoR=CoRLiking+CoRThought+CorPhysical Engagement+ . . .
This full CoR value over multiple vectors of physiological responses, which are a more complete version of the measure, takes into account the different dimensions of the media. If the media is thought provoking, creates the same positive and negative responses, and engages viewers in the same way, the media will be powerful and will do well in the market. The reason to look at a specific subset of individual vectors is to measure dimensions of the experiences of the viewers. If every viewer thinks at the same point (event) in the media, but has different emotional liking values, the conclusion is that while the media made viewers think, their responses to the media were very different emotionally, with some having a positive and some a negative response.
In some embodiments, the coherence module is further operable to group the plurality of viewers by one or more of race, gender, age, demographics, income, habits, and interests and associate and/or compare the rating of the specific one event and/or the media according to such grouping of the plurality of viewers.
There are many ways to view these measures/ratings of events and media, from a list of CoR values, to graphs, to scatter plots that include multiple CoR dimensions and/or multiple events of the media whose CoR values are plotted. As a non-limiting example,
One embodiment may be implemented using a conventional general purpose or a specialized digital computer or microprocessor(s) programmed according to the teachings of the present disclosure, as will be apparent to those skilled in the computer art. Appropriate software coding can readily be prepared by skilled programmers based on the teachings of the present disclosure, as will be apparent to those skilled in the software art. The invention may also be implemented by the preparation of integrated circuits or by interconnecting an appropriate network of conventional component circuits, as will be readily apparent to those skilled in the art.
One embodiment includes a computer program product which is a machine readable medium (media) having instructions stored thereon/in which can be used to program one or more computing devices to perform any of the features presented herein. The machine readable medium can include, but is not limited to, one or more types of disks including floppy disks, optical discs, DVD, CD-ROMs, micro drive, and magneto-optical disks, ROMs, RAMs, EPROMs, EEPROMs, DRAMs, VRAMs, flash memory devices, magnetic or optical cards, nanosystems (including molecular memory ICs), or any type of media or device suitable for storing instructions and/or data. Stored on any one of the computer readable medium (media), the present invention includes software for controlling both the hardware of the general purpose/specialized computer or microprocessor, and for enabling the computer or microprocessor to interact with a human viewer or other mechanism utilizing the results of the present invention. Such software may include, but is not limited to, device drivers, operating systems, execution environments/containers, and applications.
The foregoing description of the preferred embodiments of the present invention has been provided for the purposes of illustration and description. It is not intended to be exhaustive or to limit the invention to the precise forms disclosed. Many modifications and variations will be apparent to the practitioner skilled in the art. Particularly, while the concept “module” is used in the embodiments of the systems and methods described above, it will be evident that such concept can be interchangeably used with equivalent concepts such as, class, method, type, interface, bean, component, object model, and other suitable concepts. Embodiments were chosen and described in order to best describe the principles of the invention and its practical application, thereby enabling others skilled in the art to understand the invention, the various embodiments and with various modifications that are suited to the particular use contemplated. It is intended that the scope of the invention be defined by the following claims and their equivalents.
This application claims priority to U.S. Provisional Patent Application No. 60/905,184, filed Mar. 7, 2007, and entitled “A Method And System For Using Coherence Of Biological Responses As A Measure Of Performance Of A Section Of Media,” by Hans C. Lee, et. al., and is hereby incorporated herein by reference. This application is related to and cross-references U.S. Patent Application Ser. No. ______, filed concurrently, and entitled “A Method And System For Rating Media And Events In Media Based On Physiological Data,” by Hans C. Lee, et. al., the contents of which application are hereby incorporated by reference.
| Number | Date | Country | |
|---|---|---|---|
| 60905184 | Mar 2007 | US |
