Fundamentals of Eye Tracking in Cognitive Psychology and User Experience

Hello and welcome to the course basics of experimental design for cognitive psychology. I am Dr. Arwmer from the

department of cognitive science at IT Kpur. Uh this is the week seven of the course and we will be talking about

eyetracking in these two lectures. Now eyetracking is a very interesting methodology. Eye tracking is basically

one that helps uh researchers track visual attention. How do they track visual attention? They track visual

attention by tracking the movement of the eyes. So wherever my eyes are going, it is implicitly assumed that my visual

attention is there itself. So it helps us eye itracking helps us to detect where users are looking at a given point

in time. uh how long they are looking at that point in time uh and what is what is the scan path that their eyes have

followed. All of these factors basically are studied via eyetracking. And as we have been seeing so far, you can ask

research questions and you can ask research questions contingent to the movement of eyes. And the eyetracking

methodology will help you sort of understand how do you know uh you know get the closer to your to the answer of

your research questions by tracking how a participant is moving their eyes in a defined visual field.

The fundamental hypothesis behind uh the eyetracking methodology is referred to as the I mind hypothesis. This I mind

hypothesis was postulated by uh just enc carpenter 1980 which basically posits that there is no significant lag between

the location at which the eye is looking the aspect of uh uh you know uh that location that we are cognitively

processing. So the idea is that whatever a person is fixating at whatever wherever my eyes are resting for example

I'm reading uh from a screen and my eye is resting on a particular word on the screen it will basically mean or at

least the eyetracking methodology basically assumes that the point at which in the screen I am looking at is

the point at from which I am deriving maximum information from and that is the point which I am processing in my head.

So for example, if I'm reading a particular sentence, eye tracking is a window to human

cognition. Now when I'm looking at the screen, when I'm basically reading from here, I am basically uh let's say

eyetracking as a window. Now if I'm fixating at the window, the window is the word that my eyes are processing and

the windows is window is the word that I'm cognitively evaluating and integrating with the rest of the

sentence. So that is basically the I I mind hypothesis which says whatever you are looking at is what you are

processing and with this assumption basically is uh you know the eye tracking methodology is driven and that

is basically why they uh uh you know treat eye movements as real windows into human cognition. So the idea is wherever

we are looking we are processing some aspect of that uh field. Maybe it is an object, it is a word, it is a scene, it

is a pattern. We are processing that aspect of the given scene or a pattern in our heads. And by uh observing where

our eyes are going in a defined visual field, we can estimate what we might be processing in our heads. So that is

basically the fundamental thing that you need to know about why or how it tracking methodology is structured.

Now eyetracking is is a fairly useful uh methodology. It has been applied to several fields. It has been applied to

human factors. It has been applied to various uh topics within cognitive psychology. It has been applied to

marketing and also the broad field of human computer interaction. For example, in human factors, if uh you are a pilot

uh you know or you are a researcher who wants to optimize the uh structure of the cockpit for uh pilots, uh you might

uh you know uh put an eye tracker there and you might want to observe where does the pilot look when he first sits in the

cockpit. When he first sits uh on on this pilot chair, what are the things that are most easily accessible to the

eyes? What are the things in what order does the pilot scan the cockpit? uh what are the visual features of items

arranged in the cockpit? Are they arranged sequentially? Are they arranged logically? Uh are there uh you know some

specific buttons that have to be in certain color that should uh you know grab their attention. Basically when you

want to uh you know uh optimize where people are looking and how people are looking that is basically what you want

to sort of uh uh figure out. All right. So uh that is that is where uh you know eyetracking becomes extremely important

when you're talking about human factors even in driving for that matter. Nowadays most cars are coming with a lot

of uh you know interact interfaces and so on. Now somebody might be curious as to how does this really impact drivers

uh attention within the car and outside the car and does it dist act as a disturbance or not. So you might want to

study what are the aspects of the uh you know uh driving chamber that may distract the driver and you might want

to sort of control for them. You might want to sort of optimize them in a certain method. All right. And within

cognitive psychology say for example uh I will take examples of reading in the next lecture. So reading is something or

uh you know visual word recognition is something where eye tracking has been extremely useful. Uh also for example if

you want to study scene perception, if you want to study change detection, if you want to study how attention moves

from one part of a visual field to other all of these uh uh you know uh studies all of these topics uh eyetracking has

been uh you know an extremely important method marketing for that matter. Say for example you want to study which uh

uh you know uh shelves in the aisle uh grab most attention nowadays. uh if you see uh in big supermarkets like the

Walmarts and the uh you know uh at least in the Indian context say for example uh you have these big supermarkets uh

Reliance M and so on uh Spencers and uh you know so many other things nowadays there is a lot of supermarkets now if

you want to study for example that where do I place my product which shelf do I place my product so that it grabs

immediate attention of the person when they enter in the uh supermarket you might want to use an eye tracking study.

You might say, for example, you use a headmounted eye tracker and the participant wearing the eye tracker

walks in and is scanning the field and it gets oh this is the shelf that uh grabs maximum eyeballs and you might

want to sort of uh you know position your product in that shelf. Nowadays in in the US and so on uh the prices of uh

goods kept in the shelf or say for example for the sellers are decided on the basis of which shelves actually grab

most eyeballs. So eyetracking therefore has a lot of potential uses. It's used extensively in not only cognitive

psychology but human factors but marketing in human computer interaction. Say for example how does one interact

with a web page? What are the aspects of a web page that a person first looks at those kinds of things and in that sense

it is extremely useful and extremely uh you know important methodology. uh also besides cognitive psychology in user

experience research eyetracking helps researchers understand the complete user experience even sometimes things that

the users cannot verbalize themselves how does the user enter a space uh you know what does he look around how does

that space make him feel all of those factors say for example you are interacting with a gadget you are

interacting with a particular office space you are interacting with a particular web space how what are the

features of that web space that are appealing that are intuitive and interactive and uh which are not. So

these are some of the areas where people have used eyetracking very successfully and uh eyetracking therefore has become

a very uh interesting potential method for research uh for a lot of uh uh fields as I just mentioned. What is this

eye tracker? An eye tracker is basically a tool that allows user experience researchers or cognitive psychologists

in general to observe the position of the eyes, how the eyes are moving in the visual field and basically understand as

I said using the eye mind hypothesis, what is it that the users might be processing at a given point in time. Uh

most modern eye trackers basically rely on a method called corial reflection to detect and track the location of eyes at

it moves. So basically what happens is uh this method of corial reflection there is an infrared camera in the uh in

a you know uh in the eye tracker and that infrared camera basically sends some uh rays to the eyes and they get

reflected off the eyes and they basically track where the eyes are moving. So cornal reflection uses just

what I was just telling cornal reflection uses a light source to illuminate the eye which then causes a

reflection that is detected by this high resolution camera. So these high resolution cameras basically study how

your eyes are moving by reflecting off a beam of light of them. The image that is captured by the camera is then used to

identify the reflection of the light source on the cornea and in the pupil. And there are uh several advanced image

processing algorithms that are then used to establish the point of gaze relative to the eye and the stimula. So where is

the eye looking? Where is the stimulus located? All of those calculations sort of move around there.

This is a typical setup. You can see there is a eye tracker uh placed here. The person is looking at their screen.

Uh so uh basically as and when the person's eyes will scan the screen, you will note that uh what part of the

screen a person is looking at at a given point in time. Remember we were talking about online methods in the previous

class. Eye tracking is also an online method because you see the movement of eyes in real time and you are basically

supposed to decipher what the participant is processing in real time as their eyes are moving and scanning

across different areas of the visual field. You can see here this is basically the infrared light that is

sent off the camera. It bounces off the uh cornea and it basically falls on the screen and the movement of this uh you

know uh reflection of light is what these cameras capture. Now uh a little bit about eyes uh per se

I our eyes are typically con you know constantly moving around uh the visual field to help construct a complete

picture of what an individual is looking at. Say for example if you're scanning a a screen or if you're scanning a space

you're looking at a scene or say for example at this point I'm looking at this room I'm looking at different

aspects of this room my eyes will be moving at different aspects and basically what they are going to sort of

tell me they are going to provide me information about the environment if I suppose start to walk uh my eyes will

provide me the depth information how how long a step I have to take and so on whether I will bump into something or

not so eyes are constantly moving to uh you know derive information out of the environment and that information is

basically what I am processing and there are certain types of uh you know eye movements that are considered extremely

important some of them I will cover today some of them I'll cover when I'm talking more about reading so a fixation

is the pause of the eye movement on a very specific area of the visual flea these pauses are extremely brief 150 to

200 milliseconds at max as the eyes are continuously they are resting someplace and they are moving and they are resting

someplace and they are moving okay so when they are resting at a place that resting moment. The moment where the

eyes are resting and getting most information out of the stimulus is called fixation. All right. Now the

movement part when the eyes are moving from point A to point B that movement from point A to point B is called a

sakad. All right. So sakads are rapid movements of the eye from one fixation point to another and they help the eye

piece together a complete scene of what an individual is looking at. So for example, if there is a sentence in the

uh you know visual field and I'm reading a sentence Rajes went to buy a mango and my eyes are calling Rajesh went to buy a

mango that is basically how by moving my eyes by making these rapid sakars I'm constructing the whole sentence and I'm

integrating and understanding the meaning of the whole sentence. Now fixations basically one of the important

functions of the fixations is to basically get maximum high resolution information and that highest resolution

information comes from the phobia which is a small 2.5° region in the retina where there's the highest concentration

of the cones which provide highest resolution vision. So fixations take place in our phobial vision which

accounts for nearly half of the visual information that is sent to the brain. So maximum information uh that our brain

receives is from the phobia and there is some little information in the paraphobia and in the peripheral vision.

This part of our vision is highly detailed the uh phobial vision and it provides complete clarity very high

resolution about what an individual is actually looking at. Uh our primary attention is usually focused in uh in

what is one registering in the fial vision and this basically accounts for around 8 to 10% of our visual field

because our visual field is extremely wide. All right. So rest of our vision is composed of paraphobia region just

around the phobia and periphell region outside the paraphobia. These regions surround our phobial vision and they

help us gain a sense of what is happening around us. In this example you can see our phobial region is this small

region around 1 to 2° of visual angle. Then our paraphobial region is around 2 to 5° around the uh phobia. And then the

rest of the scene is our peripheral vision. You can see that the resolution of information that we are gaining is uh

gradually reducing. It is at peak on the phobia. Then it reduces in the paraphobia and it goes uh much lower in

the peripheral vision. All right. So these types of vision are almost certainly the results of volition where

we are actually moving our eyes and we are deciding to guess get most information out of that region. they are

also in that sense result of uh human evolution because our ancestors needed to worry about potential predators from

the corners of the eye. So that is why we also have a very wide almost uh you know 160 170° peripheral vision. Okay,

220° the maximum peripheral vision is around 220° because we want to be our ancestors probably wanted to be alert

about possible predators who are coming from all directions. Now stimuli that is registered either in the paraphobia or

peripheral vision tend to be the things that involve a lot of movement. So typically movement and uh you know

moving objects etc typically get registered in the paraphobia and the peripheral vision detected out of this

uh phobial vision uh are uh relatively low resolution and they will only give us a sense of what an object is. They

are very difficult to uh identify and they're very difficult to get the details out of. So that is something

that we have to uh you know I I'll revisit again that whatever we are seeing in the phobia this is 2° or so is

the highest resolution and the most informative part of the visual field. Paraphobia is slightly less around 5 7°

uh you know uh less detailed you can know that there is something there but you might not be able to know exactly

what is it that is there and exactly the details of that. And then there is the peripheral vision where you have a sense

of that there are things around what you are looking at but if somebody is uh is you know going to ask you the extreme

details about those objects you might not be able to provide them. You will be able to provide them those details once

you move your eyes uh to cover them bring them into fial vision and then extract most information about them. Now

while individuals obviously they cannot detect the details of objects in these uh paraphobial and peripheral regions of

our brain uh uh our brain is pretty good at uh you know anticipating and theorizing about what these objects in

the paraphobial region or in the peripheral region could be about. All right. This ability is extremely useful

in anticipating for example text when we are reading and it provides us the ability to for uh for example scan a

website uh and get the lay of the land get the overall layout of how the web page is structured within just a few

moments without even carefully reading all of the content uh you know so uh when you're scanning a visual field you

are basically it's only three four places where you actually fixate and get information out from and the rest uh of

the map of that entire field is built out of your paraphobial and peripheral vision. So human mind in that sense is

extremely uh you know capable. It is able to compose a high level understanding of a scene within few

seconds by just a few eye movements scanning the visual field. Now the unique ability of the eye

tracking uh method uh you know to detect the and follow the eyes as it looks at stimula is something that is very uh you

know useful for the user experience researchers and uh designers uh you know because it equips them with a much

better understanding of how the human visual system works. How does an individual extract information out of a

given scene or a visual construction? An eye tracker therefore can be a very powerful tool that gives us a very

highly accurate uh representation and it gives us a very detailed understanding of an individual's eye movement

behavior. And there are three very important attributes that we are gaining out of the eye tracker's reading. First

is the location of the eye movements. Where are the eyes uh you know resting at at given points in time? the duration

that the eye spends uh uh there. Say for example my eyes fixated on a particular word for uh 150 milliseconds, sometimes

300 millconds, sometimes 500 milliseconds. If the word is extremely difficult word and the movement say for

example what is the pattern of movement that my eyes are following. These form the basis of understanding how uh you

know I am grabbing or I am analyzing the information in a given visual field. The location of a user's eye gaze, you

know, where the user is looking at in in a different uh visual field at a particular point in time provides the

most basic unit of analysis and that is basically uh for understanding visual attention. So it basically gives us

where do the eyes land first, where do they go next, what is the information that they gain, how long do they rest on

their first fixation, how long do they rest on their second fixation. Those kinds of things uh basically give us a

good understanding of how our eyes are extracting information where our visual attention and how our visual attention

is moving within that given web space or within that given field of text and so on. Fixations, as I just said, are

extremely short and they're typically lasting only between 100 to 600 milliseconds. And they can uh, you know,

be mapped on between specific X and Y coordinates on a grid that can help pinpoint where exactly your eyes were

resting, where exactly the user was paying attention to given a particular display. The challenges however with

interpreting mapped fixations lie in the fact that just because the fixation was registered does not necessarily mean

that the user really saw it and uh that it was registered cognitively in the brain. So that is a little bit of a

trade-off that one has to really uh pay attention to. This is often the case with sometimes you know orphan fixations

when you are randomly scanning the screen where the eye may have momentarily rested in a random area of

the screen but it did not intentionally extract information out of it. they did not intentionally look at it. Such as

when users are say you know completing forms that require created answers in such cases users are looking at the

screen but their attention is somewhere else. They're actually thinking of what will I fill in this form? What kind of

answer do I build up? So in those cases you have to be very careful that the eye tracking data may not be extremely

useful. Now the clustering of I fixations where the eye is I fixations are resting which

area is considered most informative uh for the individual in a particular region that can actually provide us

more evidence that the user is deliberately looking at something and trying to access information out of that

region and significantly increases the likelihood that the brain has processed that particular region. All right. So if

there are several fixations around a particular scene around a particular aspect of a web page, you might and be

sure that that is the region that the uh human brain is processing in some detail. Uh it trackers uh in that sense

are not really mind readading devices but they can only tell us what the person is looking at. Uh they don't tell

us why. So you have to basically create your hypothesis in such a way that you know that okay what was there in that

region. For example, in uh text reading, uh sometimes eyes will rest at an ambiguous word. So for example, the

horse that the horse that raised past the barn fell. Now if that is a sentence and the eyes are coming back and resting

at the barn again and again, you know that there is some ambiguity in that sentence structure that the eyes are

trying to resolve and that is why they are coming back. They are fixiating there. Sometimes there are regressive

eye movements. Eyes move forward and they come back and they fixate there again because they're trying to get

information out of there. they are trying to say resolve oh what does this sentence mean all right much more of

this I will discuss when in the next lecture when I'm talking the ex taking the example of reading an eye moments

now the length of the time that a user fixates on a particular area on a screen helps us to understand whether he or she

is paying particular attention at a given uh specific visual element in the scene given specific visual element of

the web page and that is basically something that is uh informative for designing researchers, for user

experience researchers, for human computer uh computer interaction researchers and also for cognitive

psychology researchers. Okay. Now the duration of fixations, how long did the eye rest at a given place is also

typically extremely short and is represented only in milliseconds, few hundred milliseconds that is. Now while

duration is a particularly difficult measurement to interpret because you cannot really know why did a user fixate

for such a short time for that or a long period of time on that area. You do not know why. So you have to create the

environment such that you know okay uh if the eyes are fixating here this is probably the reason why uh did they find

it engaging or they would be confused by relevant those kind of questions you have to uh you know handle in your

experiment design. This is for example an uh you know uh a map of how I fixations are moving on a

web page. You can see the eyes are landing here and then they are going there and there and there and so on. So

basically you can see the eyes uh or the fixations are mapped on an xy. So this is the x this is the y are mapped on an

xy coordinate in the grid and that basically tells you the precise location where the eye is landed at. And this

basically can tell you how the individual who's looking here is scanning the screen is reading a

particular website. Here you can see uh uh again this is a typical scan path. There is the you know

number of fixations 1 2 3 4 5 6. You can see here interestingly uh the longer and shorter durations of how much time did

the eye spend is represented by the size of the circles. The relative size of the dot. So where the eye spend more time is

actually uh a larger dot where the eye spend very little time is represented by a smaller dot. So the time period here

is represented by the amount by uh you know uh the size of the dot. If the eye is pres uh is uh spending more time at a

particular point, the dot is uh slightly enlarged. If I is presenting is spending a very small time uh here the eyes

presented by a smaller dot. The movement of the eyes of a user's eyes is typically based on sakads. Sakads are

rapid eye movements that I just talked about from one fixation to other and it establishes the eye gaze pattern that

reflects on how the user interprets a particular visual stimulus. How is what is the strategy of the user in getting

information out of a visual stimulus? Be it a web page, be it a given sentence. How do the eyes move from point A to

point B and get maximum information out? Now, this pattern provides the basis for understanding the visual hierarchy of

the scene. How is the scene organized? What are the more important elements? What are the less important elements?

How does the user begin to sample that particular visual space? which uh uh you know spaces in that visual space uh are

sampled first and then where does the eye go to where the eye goes to the third uh place or the fourth place where

does the eye spend most of the time these are the things that as experimenters as researchers you are

actually concerned uh with so visual hierarchy as I was just saying refers to the sequence in which a user views uh

visual elements in a given scene for example on a website a user might first notice a large graphic in the center of

the screen. Maybe it's the heading of some kind uh or the uh title of the page. Then they look at the primary

navigation tools. How do I go out from here? How do I change the page? Uh and so on and then looks at the search box.

Oh, I have to look for something. Say for example, a typical Google web page. So the eyes probably first land at the

big Google uh trademark that is there and then they go onto the search space and you want to type something and get

information out of that. Okay. So eyetracking in that sense is particularly good at revealing the

effectiveness of how the organization how the gestal principles of design influence the order of elements that a

user is looking at. So in that that is basically how and uh why you know lot of researchers especially UX designers and

especially HCI people are interested in working with the eyes marketing people say for example you have a very good

online uh website uh you know let's say an Amazon or something like that you want to understand where does the user

go where does he land his eyes first how does he scan the uh arrangement of uh products uh are the arrangements uh you

know exactly appealing or not. Are they grabbing attention or not? And what is the order in which the user is scanning

your web page? All of this information becomes extremely useful and it becomes uh you know informative for uh designers

of the web page to optimize to grab most attention uh to uh get the best out of their uh you know prospective clients

and in their interactions with these web pages. So this is again uh an example of

eyetracking visualization. This is the scene where uh you know which is there and you can see when you give free

examination people are sort of randomly going through when you uh ask people to estimate the material circumstances of

the family tell whether the family is rich or poor. Then you can see where they are looking at what they are

looking at the paintings probably they're looking at the dining table and so on. Uh when you ask them to estimate

the ages of the people then they're looking at these four people present five people present in the thing. So

they're looking at the people and they are getting most information out of them. How old a given person is. Surma

is what the family had been doing before the arrival of this visitor. So they are basically looking at the arrangement of

the room. Uh uh you know remember the clothes worn by the people. So they are again looking at the people. They're

trying to get what the everybody is wearing. So you can see here that eye movements are extremely

responsive to task instructions. What is it that you are asking your participant to do? Are you asking them uh you know

to just read a given sentence? Are you asking them to derive the meaning of individual words of a sentence? Are you

asking them to scan the scene and get specific information out? Say for example here participants were asked to

give the ages of the people. So you can see here the eye movements are actually focused on 1 2 3 and then four and five

people who are present in the scene not on objects but on actual persons. When you are asking them to look at the

material circumstances of family then they are looking at the paintings they're looking at the furniture and

they're trying to get oh how expensive or inexpensive this furniture is which will give us an idea about how are the

material circumstances of the family. Again you can see remember positions of people is an object in the room. There

you can see wider scanning of the room uh tagging of people tagging of placement of furniture in the room and

so on. So uh when you're designing experiments uh you know using eye movements it is extremely important to

understand what is it that you want to uh you know get out of these eye movements. What is the question? So that

is why remember in experimental design it is extremely important to know what your research question is. Your research

question should mostly determine the method that you're using and it should not be the method that oh I will use

eyetracking in my experiment and then worry about what is it that I want to get out of eye tracking. The order

should always be this is my research question. Uh I am designing uh these experiments and a behavioral experiment

or an IT tracking experiment is going to answer my research question in this particular way. That is basically the

order in which one should actually move. Okay. Now nowadays there are both hardware and software uh that are

available to both researcher and the participant. Uh they make the job of uh you know extracting eye movements and

analyzing eye movements much easier. researchers can almost seamlessly integrate eyetracking into their user

centered design in their cognitive psychology experiments with minimal uh you know uh issues. Uh for example,

advancement in remote eye tracker technology can now make it possible to calibrate the uh you know movement of

the eyes of a uh you know user very quickly. Eye trackers of the day are nowadays extremely accurate. They can

track diverse kinds of population even uh populations with uh issues for example children or older people. uh

their eyes can also be tracked relatively easily. Now the IT is used today uh come with software suites uh

for example in our lab we use 10,00 plus SR research IT tracker. It comes with its own experiment builder. It comes

with its own data extraction software which allows the students to quickly learn and pick up uh you know how they

will extract uh information out of the IT tracker and get design their experiments out of it. The output from

these software packages helps to highlight uh where the user is looking at what is the length of the time they

look there and their gaze patterns uh and how did they sort of move across a given scheme. Some of the most commonly

used visualizations that might be extremely useful are say for example the heat map and the gaze plot. So heat map

is a visualization that uses different colors to show the amount of fixation. How many times a person fixated at a

given point in time and for how long they fixated at these areas, what time did they spend uh you know how much

fixations and how long the fixations were for. Heat maps are colorcoded. So red is typically used to indicate a

relatively higher number of fixations or duration and the green uh basically indicates the least number of fixations

and fixation durations. uh and then there is a overall uh you know uh gradient of colors in between. An area

with no color on the heat map signifies that the participants may not have looked at or fixated at that area and it

does not necessarily mean that they did not see anything but they just did not rest their eyes on that area. Okay, so

they may look may have looked at there for a short period or may have only registered that uh region peripherally

uh but it may not have been detected by the eye tracker. All right. So this is for example uh you know a se heat map of

an individual looking at a particular scene. So you can see uh the face is where the people uh pay most attention

to and then they pay most attention to this uh uh surfing or diving individual and the other face and and so on. Uh

there are other places where the users have probably not really looked at. Gaze plots are visual representation of

fixations and zakads for a particular time frame within a particular time. What is the gaze plot? How did the eyes

move from point A to point B in that visual field? In most software applications, uh you'll find that the

fixations are represented by dots and secards are represented by lines connecting those dots. I'll just show

you uh very quickly. Uh fixations are numbered to show the order of fixations which fixation happened first. So 1 2 3

4 and so on. Uh and can be varied in size to illustrate the time that was spent on each fixation. So there is one

so we saw heat map then there is gaze plot then the other thing is area of interest. Uh sometimes when you have a

particular hypothesis you want to know whether my participant looked at this uh particular area or not then what you can

do is you can divide your region and this is especially done in sentence reading studies is that you divide your

uh sentence into areas of interest and then you're specifically looking for fixations landing in those areas of

interest. Okay. So areas of interest typically help the researchers analyze the various components of a

visual scene. Uh researchers can basically categorize regions of a display into different geometric shapes.

Sometimes they are circular, sometimes they are rectangular that correspond to certain elements on the screen. So for

example, when you're uh looking at a sentence, typically rectangular uh blocks can be made. So in analyzing the

various components of the website, the researcher may want to create AOIS uh these areas of interest for the primary

and secondary navigation, the search box, graphical and textual elements and basically look at whether the user's

eyes actually landed at those uh places or not. uh the data from the AOIS then can be aggregated and they can be

aggregated across several participants to understand the order in which all participants uh scanned the website

which areas of interest received most prominent attention which areas of interest did not receive a lot of

attention which were appealing which were not appealing. Sometimes you uh you know design your graphical elements such

that they grab attention but uh during this analysis if you know that oh this did not really uh grab a lot of people's

attention you might want to change the color scheme you might want to change the design arrangement and so on. So

these analysis basically uh you know make it relatively easy to analyze the results from eye tracking study. You can

look at this say for example these are again heat maps uh and these heat maps are basically telling us how the eyes

are moving across these given web pages. This is uh the gaze plot. You can see here uh you know eyes are moving from 1

2 3 4 different regions. And the size of these dots basically represent how how much longer the eyes actually looked at

uh at a given position. So this is one somewhere it is two this is three uh then there is uh four and this is

basically the scan path of how the eye is scanning that given web page of healthcare.gov.

So this is all about eyetracking and user experience in cognitive psychology. In the next lecture I will talk about

eyetracking with a special focus on reading studies. All right. Thank you.