- #WE WERE HERE TOGETHER WALKTHROUGH CHAPTER 10 SOFTWARE#
- #WE WERE HERE TOGETHER WALKTHROUGH CHAPTER 10 TRIAL#
#WE WERE HERE TOGETHER WALKTHROUGH CHAPTER 10 SOFTWARE#
I would recommend that you set up your experiment software so that when the participants press the Z key, it sends a message at the same time saying ‘RESPONSE_PRESENT’ and so that when participants press the M key, it sends a message at the same time saying ‘RESPONSE_ABSENT’. Suppose that participants here are asked to press the Z key for one response to indicate that they have found a target and the M key for a second response to indicate that they think no targets are present.
![we were here together walkthrough chapter 10 we were here together walkthrough chapter 10](https://gameplay.tips/uploads/posts/2019-10/medium/1570985188_35.jpg)
To begin with, what we need to have is a message sent to the message report detailing two things: the time of the response, and the response itself. There are other ways to achieve this, but here is what I would recommend to go for since the approach suggested below fits in with the expected inputs and outputs of other functions that form part of eyeTrackR. This is the case if participants responded using a keyboard or a mouse. If your participants didn’t use a standard button box, then you need to use a slightly different approach. Here, since the button boxes you can connect to Experiment Builder get given numbers (1-8), the correct response lists the correct numerical value of the button that needs to be pressed on each trial.
#WE WERE HERE TOGETHER WALKTHROUGH CHAPTER 10 TRIAL#
This function needs to be fed the fixation report you are using, as well as the column which states what the correct response on each trial is. To mark up the fixation report with the timings of each response, as well as the button press of each response, and finally the outcome (correct or incorrect) of each response, we can use. If you are not using a button box, you’ll need to use the method described in Behavioural Markup (Alternative). If you use a button box, you can use the method here. Here, there are two ways to markup the fixation report. The only thing that is missing is the other main event in each trial: the response made by each participant! We now have two of the major events in each trial marked up into our fixation report. If you get the following message, then you have nothing to worry about because the dataset has been matched up perfectly:ĭifference between input and output rows: 0
![we were here together walkthrough chapter 10 we were here together walkthrough chapter 10](https://venturebeat.com/wp-content/uploads/2018/04/bc-og-940x529-hero.jpg)
Usually this is the case that you have got participants or trials that are missing from either the fixation report or the message report. If the number of rows are different, that suggests something has gone wrong. It compares the number of rows your fixation report had before you marked it up with each new message with what happens to the fixation report after you have marked it up with each new message. This process also gives you some diagnostic information that helps to check for any problems. Take the timings of those messages and then add that information to the fixation report.Work out when each DISPLAY_START and DISPLAY_CHANGE occurred in each of the trials.If you look closely, what this has done is 5: One is called DISPLAY_START and the other is called DISPLAY_CHANGE. If you now take a look at fixationreport - which I would recommend you do - you’ll now see some new columns have been added. # MARKUP FIXATION REPORT fixationreport <- ( message_df= messagereport, fixreport_df = fixationreport, message= "DISPLAY_START") fixationreport <- ( message_df= messagereport, fixreport_df = fixationreport, message= "DISPLAY_CHANGE") Now that you have checked to make sure things appear as they should - here, we have XYZ participants in our example dataset who took part in XYZ trials, giving us XYZ trials across everyone, which is what is being reported in the table, we can continue.
![we were here together walkthrough chapter 10 we were here together walkthrough chapter 10](https://i.ytimg.com/vi/uCPqQcHpi20/maxresdefault.jpg)
If that’s not the case, then you may need to delve deeper to find why data are missing. Assuming you know how many participants were in your dataset, and how many trials, you should be able to check to make sure the right number of trials have started (i.e., had a DISPLAY_START message appear).
![we were here together walkthrough chapter 10 we were here together walkthrough chapter 10](https://cdn.cloudflare.steamstatic.com/steam/apps/865360/ss_b034e5f22194605fc81203c63432337f351e6237.1920x1080.jpg)
One thing you can do at this point is a back-of-the-envelope calculation and check to make sure all the trials are there as you would expect. There will always be variability in terms of when events occur in a trial, but this does enable you to make sure that, for the most part at least, things happened when they were supposed to. Then, it gives you a count of how many times each message has been detected in your message report, the mean time each message occurred from the start of each trial, and then, as well as that, the maximum and minimum times each message occurred. The table above, which you get in the console output by running, has some useful information in it that you can use to spot and diagnose any problems with your data.įirst, it lists all of the messages it finds in your message report file, in turn, and puts them in the CURRENT_MSG_TEXT column. # TAKE A LOOK print( (messagereport))įrom that, we get the following - formatted nicely here for easy reading: Table 4.1: Message report descriptives