Possible Project Topics

Listed below are some potential domains within which you might specify your project. They are not, in themselves, project specifications. Each topic could be tackled in various ways, and final projects could involve a combination of elements, such as modelling and experiment, or phenomenological analysis and theoretical critique.

Based on the expressed preferences of students, you may work individually or in groups of up to 4 students.

You are free, nay encouraged, to develop your own research question. Do discuss it with us first though.

Project Topics

1. Earworms: the tunes that get stuck in your head.

Everyone is familiar with the phenomenon of a tune that seems to be stuck on endless replay. One conventionally describes this as happening "in the head". Indeed, given that the repetition seems to continue without any necessary input from the environment, it is plausible that there is a well-defined neural process underlying this, though very little is actually understood about how this happens, or why. There are many different approaches to this, of varying promise and quality, and they do not all cite or refer to each other.

An initial set of questions revolve around the phenomenology of an earworm. Does everybody experience this in the same manner? Does it bear similarity to voices in the head? Is it similarly intrusive for everybody? What strategies for quenching or replacing the earworm exist, and do they work? Do they do so for everybody? How does it interact with cognitive load? If you were to model this, which aspects would you model?


2. Visual Arithmetic

People seem able to rapidly estimate the approximate magnitude of answers of 'visual sums' fairly well. For example, if you briefly show an image with 10 dots, a plus symbol, 15 dots, an equal symbol and something around 25 dots, they will quickly respond by indicating the sum is correct; if you show an image with 10 dots, a plus 15 dots, an equals and something around 40 dots, they will quickly respond by indicating the sum is incorrect. The aim in this project will be to try and answer various questions about this "magnitude estimation" or "visual arithmetic" ability. Some questions could be: how accurate are people's magnitude estimates? How fast can this process be? Is it affected by properties of the dots used (do the dots all have to look the same)? Is it a similar process to rapid magnitude estimation for actual numeric sums (where people are shown numbers, not dots)? Can people do this magnitude estimation for multiplication? For division?


3. Eye movements and moving images

A lot of work in visual perception has relied heavily on viewing of static images. But the world is not static, and unmoving images are a rather odd type of artefact, rather than a paradigmatic instance of a visual stimulus. The study of how eyes move introduces possible overlaps between cognitive science and media studies. Hitchcock famously said that he liked to play his audience like the conductor of an orchestra. In the link below, it was demonstrated that some kind of cortical activity is similar across subjects as they watch films. But just as interestingly, eye movements were highly similar across subjects. The relation between synchronzised eye movements and film type, purpose and meaning is a topic ripe for exploring.

Tech note: we have a 60 Hz eye tracker available that will record eye movements as material is presented on screen. Expect to spend a bit of time learning how to set up experiments, record data, export it, and analyse it. Creativity in analysis will be necessary.


4. The Endowment Effect

The endowment effect is a hypothesis in behavioural economics that people ascribe more value to things merely because they own them. An example of this is a wine collector who will only buy a certain type of wine X when it's price is 50 dollars or less. When that X wine goes above 50 dollars in price, the collector refuses to buy any more (indicating that they don't think it is worth more than 50 dollars). However, when a professional wine-buyer offers to buy a bottle of X wine from the collector for 100 dollars, the collector refuses to sell (indicating that they think the wine is worth more than 100 dollars). The standard explanation for this pattern is that, for the collector, the wine is "worth more" simply because they own it. The aim in this project is to test this explanation of the effect by comparing it against alternative explanations. One such alternative explanation is that the effect arises because, just by offering to buy the wine for 100 dollars, the buyer is telling the collector that the wine is actually worth 100 dollars or more, and so causing a change in the collectors valuation of the wine.


5. Gaze and Conversation

Observation is a learned skill. Part of the skill of observation finds expression in experimental science as the coding of complex attributes in the annotation of complex behaviour. Coding and annotation of video material can inform us greatly about the manner in which subjects become entwined during face to face conversation. Some corpus data is available for study, or students might record their own. Body language, discourse moves, attitudes, conversational moves, gaze and blinks, are all examples of the kind of complex attribute that can be annotated in coming to understand conversation.


6. Illusory Correlation

Illusory correlation is the phenomenon of perceiving a relationship between variables (typically people, events, or behaviors) even when no such relationship exists. A common example of this phenomenon would be when people form false associations between membership in a statistical minority group and rare (typically negative) behaviors as variables that are novel or salient tend to capture the attention. Illusory correlation is one way in which stereotypes form. The most accepted current account says that illusory correlation occurs because people people expend more effort encoding distinctive (that is, more rarely occurring) information, and so that distinctive information is more strongly recalled. Because a minority group is by definition more rarely occurring, variables occuring with minority items are more strongly recalled and so are seen as more strongly associated with the minority group. An alternative explanation says that illusory correlation arises because random error in recall causes rare events to be seen as more likely than they really are (a statistical process called 'regression to the mean'). These two accounts make various conflicting predictions. For example, the distinctiveness account predicts that if an artificial situation is set up where members of the majority group was somehow more distinctive than the minority, then the variable of interest would be more strongly associated with the majority than the minority (even though in fact there was no association); on the other hand, the 'random error' account predicts that even in this situation, the variable of interest should still be associated with the minority. The aim in this project would be to identify such predictions and run experiments to explicitly test them.


7. Periodic and Aperiodic Synchronization

It is commonly assumed that synchronization, in the sense of doing the exact same thing at the same time, depends on the presence of a beat, or clock. Our exquisite ability to speak in synchrony demonstrates otherwise, as there is no beat in speech. Can you get people to synchronize their handwriting? How would you operationalise this? Could you use tablets, smart whiteboards, or other technological means to obtain your data? What aperiodic actions can be synchronized over long time spans? What is necessary for synchronization? How complex a gesture can be synchronized? There is much to be done here. Empirical study will demand some cleverness in obtaining appropriate data.


Some additional topics (do the leg work yourself)

The following are just seeds. Feel free to discuss any of these with either Fintan or myself.

Blinks, blinks and more blinks
Do blinks tend to synchronise among musicians playing in an ensemble? Among people singing in a choir? Among rowers in a boat? Among people listening to the same music? I don't know. Do you? How would you argue this?
The ill-defined notion of style finds application in many domains, such as writing, cooking, dressing, arguing, walking, living. Can this poorly defined notion be made precise in ways that generalise to at least two domains (e.g. writing and making music)?
So many questions. What qualitatively different kinds of laughter are there? How do you demarcate the situations in which each occurs? How can you come to grips with laughter empirically?
Excuse me? Capturing someone's attention
If one person strikes up a conversation with another who is not inviting it (beggars, advertisers, interviewers, people looking for directions), what are the structures of the brief exchange before which the two people are essentially uncoordinated, and after which they are coordinated. Can this be systematised?
Over-practiced Speech
What are the characteristics of over-practiced speech? E.g. Airline Safety Announcements, Prayers? Where does this question lead?
The skill of attention splitting
Can people learn general techniques that allow them to "split" their attention? If they learn to split their attention for one task (e.g., talking on the phone and simultaneously watching TV), does that help them split their attention in other tasks?
The skill of multi-tasking
Is there a general skill of "multi-tasking"? If some people can easily do one pair of tasks simultaneously, can those same people easily do some different pair of tasks simultaneously? Or must a "multi-tasking" ability be specifically learned for every pair of tasks?
The nature of multi-tasking
When we multi-task, are we really doing two different tasks in parallel? Or have we actually learned to carry out a single "higher-level" task that has two separate "sub-tasks" as components?
The venerable problem of insight in problem solving
"Insight" problem solving refers to the ability to solve a problem by reassessing the frame or assumptions implicit in the problem. Here's a problem requiring some form of insight: "Can a man marry his widow's sister? Why or why not?" (The insight here is: if a man has a widow, that means the man is dead; dead people can't marry). A research question related to insight might be: Is "insight problem solving" a skill that can be learned? Does "insight" improve with practice?
Critique the notion of the "natural image"
In studies of vision one frequently finds the term "natural image", often meaning a photo of a mountain landscape or similar scene. One may also find reference to a "natural movie". Critique this idea. Trace its usage. Find out what the users of such terms mean by "natural" and bring your critical faculties to bear. This will involve literature search, perhaps correspondance with scientists, and criticism. It does not require you to do an experiment. (Sample paper to get you started: Boccignone, G. (2015). Advanced statistical methods for eye movement analysis and modeling: a gentle introduction. arXiv preprint arXiv:1506.07194.)
Synchronisation in clapping
In Western Europe, when we applaud, individuals do not synchronise. In Eastern Europe, they do. Synchronisation is a sign of boredom over here and of enthusiasm over there. So what happens if a bunch of people over here are asked to synchronise their clapping? Is it easy? Do they always converge on the same rate? Could one use this as an experimental structure, with a priming event or task beforehand, and use the rate, variability, and other overt signatures of the clapping as a dependent variable? How should we interpret this remarkable cultural divide?