Conventional wisdom has claimed that choice, and the freedom associated with it, is always is good thing and the human ability to desire and manage choice is unlimited. However, recent evidence - popularized by Barry Schwartz in his popular book "Paradox of Choice" and a famous jam study by psychologists Mark Lepper and Sheena Iyengar - appears to indicate that too much choice is actually a bad thing, causing decision paralysis and unhappiness. See Barry Schwartz in this TED talk
In the jam study, when faced with choosing (tasting and buying) between jams in displays with 6 and 24 varieties of exotic and high quality jams, though more customers were attracted to the display with 24 varieties, only a small share of customers who visited the larger display actually purchased a jam whereas a much larger share of customers visiting the smaller display purchased a jam.
However, as Tim Harford writes, subsequent experiments on similar contexts (including jam and luxury chocolates) find inconclusive evidence that increasing choice demotivates consumers. He writes, "The average of all these studies suggests that offering lots of extra choices seems to make no important difference either way."
However inconclusive the evidence, I am personally inclined to accept the view that a large choice set does place constraints on human ability to make efficient and optimal choices. As to what constitutes a "large" choice set, it varies across contexts and objects being chosen. The most commonplace examples of abundance of choice leading to sub-optimal outcomes and lower happiness are selection of dress materials (say a shirt) in a large showroom (especially those with multiple brands) and choosing from an a la carte menu in an upmarket restaurant. From numerous such less than happy experiences with such choices, at both cloth showrooms and restaurants, I am personally convinced that "too much choice is bad", at least in some circumstances.
Apart from the forementioned examples, it is fair to say that when making choices on not so simple objects like insurance or savings policies and financial products, complex electronic products, etc, the revealed preferences of people indicates that more structured choice leads to more optimal and happier outcomes. As Stephen Dubner writes, "So even if jam studies of the future prove inconclusive, it still seems wise to streamline choices whose complexity might otherwise hamper a good outcome". Paternalistic "nudges" may lead to optimal and happy outcomes in such circumstances.
Benjamin Scheibehenne, Rainer Greifeneder, and Peter M. Todd conducted a meta-analysis of 50 published and unpublished experiments, depicting the choices of 5036 consumers, that investigated choice overload and found that consumers generally respond positively to having many choices. They find that the overall effect of choice overload was virtually zero.
Update 2 (13/11/2010)
Christine Benesch, Bruno S. Frey and Alois Stutzer have a new working paper which finds that "heavy TV viewers do not benefit but instead report lower life satisfaction with access to more TV channels. This finding suggests that an identifiable group of individuals experiences a self-control problem when it comes to TV viewing."