In this article you will discover what the main metrics of eye tracking research are and how to interpret them.
The best way to see through your customers’ eyes without getting in the way of their natural behaviour is using eye tracking. This methodology allows you to measure eye movements of a person when they are interacting with a visual stimulus and can offer relevant information about subconscious human behavior.
Eye tracking can be used on its own or combined with other methodologies for qualitative and quantitative research in several fields, such as market research, usability testing, academic research, gaming, in the medical context, amongst many others areas of specialty that can benefit from data provided by this technology.
Conducting an eye tracking research can be simple using the right tools, but many researchers might be overwhelmed with what to do with the amount of data that they get.
To uncomplicate it, we listed below the main metrics measured in eye tracking, their definition and how to read them in order to obtain valuable insights:
Definition: An AOI is a tool to select areas of a displayed stimulus(e.g., an image, website, video etc.), and to extract metrics specifically for those regions. Any area or object in a stimulus can be an AOI. An AOI is not a metric by itself, but defines what area you calculate further metrics for.
Definition: Time to first Fixation (TTFF) is the time from the start of a task until a particular AOI is fixated for the first time.
Interpretation: A short TTFF means that your AOI was discovered and fixated very quickly by your participants. This indicates a strong salience ofyour AOI (e.g., it’s very colorful or flashy, it’s very central in the screen,it’s big, it visually stands out compared to the other objects in some way etc.). These bottom-up characteristics of your AOI drive the very early processing (0-0.15s) which can’t be actively controlled by your participants.
A longer TTFF can indicate that your AOI is not as salient as the objects around that AOI. AOIs with a longer TTFF are most often discovered by active and conscious visual search and top-down searches of your participants (e.g., the participants look at a certain place on a website or image because they expect a certain element to be there, or the participants actively decide to focus on certain elementson a website or picture). TTFF is a basic yet very valuable metric in eye tracking, as it can provide information about how certain aspects of a visual scene are prioritized.
Definition: This is the duration of the first fixation on a particular AOI.
Interpretation: A short (0.1 sec - 0.2 sec) FFD on an AOI can indicate that your participants were distracted by other areas on screen. This can have two causes: firstly, the other areas can be more salient (e.g.,bigger, more colorful, blinking etc.) and automatically catch the participants’ attention. Secondly, the other areas could be more interesting to the participant, or be perceived as more helpful to complete their task. In that case, the participants actively decide to shift their attention to the other area. A long FFD can indicate a high level of interest or task relevance of an AOI.
This metric is especially interesting in combination with the Time to first Fixation (TTFF): a short Time to first Fixation combined with a short FFD likely means that your participants have noticed this AOI in the very early stages of processing which is driven solely by bottom-up characteristics of your AOIs. If the participants revisit this AOI they have actively decided that it is relevant or interesting to them. A long TTFF with a short FFD, however, means that your user is already in the active processing stages and has decided that your AOI doesn’t contain interesting or relevant information. However, if your AOI contains only very little information, a short FFD might be enough to process that.
Definition: The dwell time is the total time your participants spent looking at an AOI. We usually calculate the average dwell time, which tells you how long your participants looked at an AOI on average.
Interpretation: The dwell time depends on the size and information content of the AOI you’ve drawn, and it’s therefore hard to quantify how longor short a dwell time should be. It is important to keep that in mind when comparing the dwell time of AOIs.
In general, a shorter dwell time (compared to other AOIs with a similar size and information density) on an AOI can indicate that the participants actively decided to shift their attention away from an AOI to the other areas of the screen. Those other areas could have been more interesting to the participants, or been perceived as more helpful to complete their task. A long dwell time can indicate a high level of interest or task relevance of an AOI. However, if the information in an AOI is very complex or confusing to participants, they might also have a longer dwell time on that AOI.
The bottom-up characteristics of your stimulus (such as salience due to object color, movement, size etc.) have only a minor influence on the dwell time. This makes the dwell time a very insightful metric, since it can give you insights on your participants conscious eye movements and judgements.
Definition: The average fixation duration tells you how long a fixation on a certain AOI lasted on average. The main difference between Dwell Time and Fixation Duration is that the first measures all fixations cumulated and the latteris for one fixation. This metric is not only relevant in relation to the AOI, but also gives an indication of search behavior for the whole stimulus.
Fixation durations typically last from 0.15-0.3 sec.
Interpretation: A short average fixation duration can indicate that the AOI doesn’t contain a lot of (relevant) information, and the participants shift their attention towards more relevant AOIs. It can also indicate that the information in this area is particularly easy to read and understand, so the participants do not need a long time to process it.
In the very early seconds of a task, a short fixation duration can indicate that there are other AOIs around, that are more salient and therefore distracting the participants. A long fixation duration can indicate a high level of interest or task relevance of an AOI. However, if an AOI or stimulus is confusing to participants, they might also have a longer fixation duration on that AOI.
This metric is especially useful in combination with the number of fixations/revisits to an AOI. For example, if there is a lot of revisits with short fixation durations, this behavior can indicate that people are searching for something in particular but might have trouble finding it because of a high visual complexity or lack of clarity. A lower number of revisits with longer fixation durations are in general associated with a deeper processing of the information in an AOI.
Definition: This is the number of participants that fixated on a particular AOI.
Interpretation: If some of your participants missed a certain AOI and didn’t fixate on it, that likely means that it was either very hard to spot (e.g., it wasn’t salient in comparison to the other areas on the screen), or the participants didn’t search for it, because it likely contained information that they perceived as irrelevant.
Definition: Number of times a participant revisits a particular AOI.
Interpretation: A high number of revisits on an AOI can mean that it contains very task relevant or interesting information, which the participants decide to come back to frequently. It can also mean that the information is too complex, and takes a couple of reads to be fully understood. The AOI might also be confusing, or even frustrating. An object that is pleasing to look at might also have multiple revisits.
The interpretation of revisits can only be made when looking at the exact task and stimulus at hand.
Measuring eye movements can offer immeasurable value for research in different fields by showing data that would not be consciously verbalized by the participants. By knowing the metrics and how to read them, researchers can get insights into human behavior and decision making processes.
If you want to explore the power of eye tracking, contact us for a free trial of Oculid's platform.
November 17, 2021
1. Theeuwes, J., Atchley,P., & Kramer, A. F. (2000). On the time course of top-down and bottom-upcontrol of visual attention. Control of cognitive processes: Attention andperformance XVIII, 105-124.
2. Theeuwes, J., Godijn,R., & Pratt, J. (2004). A new estimation of the duration of attentionaldwell time. Psychonomic bulletin & review, 11(1), 60-64.
3. Theeuwes, J. (2010).Top–down and bottom–up control of visual selection. Acta psychologica, 135(2),77-99.