Alison Abbott in Nature:
…A typical study probing UTA asks subjects to make a complex decision, such as choosing a car or a computer, after either mulling over a list of the object’s attributes or viewing the list quickly and then engaging in a distracting activity such as a word puzzle. However, such studies have drawn different conclusions, with about half of those published so far reporting a UTA effect and the other half finding none. Proponents of the theory claim that the effect is exquisitely sensitive to experimental variations, and often attribute the negative results to the fact that many research groups varied elements of the set-up, such as the choice of puzzle used for the distraction3. Critics say that the positive results came from having too few participants in the experiments. Psychologists Mark Nieuwenstein and Hedderik van Rijn at the University of Groningen in the Netherlands set out with their colleagues to determine which explanation was correct. They asked 399 participants — around ten times more than the typical (median) sample sizes in other studies — to choose between either 4 cars or 4 apartments on the basis of 12 desirable or undesirable features. They incorporated the full list of conditions that UTA proponents had reported as yielding the strongest effect, such as the exact type of puzzle used as a distraction. They found that the distracted group was no more likely than the deliberating group to choose the most desirable item.
More here.