HCI Car Phone Interface
Research question: Do people get less distracted while driving while using the hands-free setting to talk on the phone as opposed to holding their phone in their hand?
Variables:
Independent variable: Whether the person is using the hands-free setting or holding their phone in their hand
Dependent variable: Number of cars people count as being around them on the road, speed of the car, and turn signal usage.
What I’ll do:
The plan is to use a driving simulation that will allow the user to be able to check their mirrors to see cars around them. The participant will need to count the number of cars that are on the road while answering a series of questions about themselves and while obeying the rules of the road: turning on their turn signal when turning and going the speed limit. There will be five short and simple questions, and five more complex questions to mimick the complexity of a conversation.
The same group of people will be tested over three days. The first day they will drive without talking on the phone. The second day they will drive while having their phone in their hand, and the third day they will drive using a hands-free mode.
Measurements:
The participants’ level of distractedness will be determined by how many cars they identified around them, whether they went the speed limit (+/- 5), and whether they used their turn signal for each turn. If participants are unable to do at least one of the three completely accurately, they are distracted.
Hypothesis:
I believe that the participants will be more distracted while holding their phone in their hand because there is the additional stimuli of needing to hold an object that has nothing to do with driving, whereas while using the hands-free mode there is no additional tactile stimulation.
If it is proven that participants are more distracted while holding their phone, my future work will test if this is the cause of the heightened distraction.
Nice ideas re the distracted driving study. I’m pretty sure that you could find studies out there similar to this if you want to see the results. In your particular study design, having distracted be a binary / dichotomous variable (they are distracted or they’re not) makes the statistical analysis a little more complex. You can use logistic regression and other methods, but your life would be easier if you could make it a continuous variable. Perhaps you create a composite score for distractedness that is based on a weighted function of accuracy-of-nearby-car-count, speed-limit variation, and turn signal compliance.