Three studies were conducted with the aim of investigating whether choice o
f behavioral alternative increases intention-behavior consistency. In Study
1 undergraduates were asked to write and mail back an essay on a topic the
y either chose or were assigned. Demonstrating an effect of choice, signifi
cantly more subjects who chose an essay topic mailed back the essays. These
subjects also rated their intention to write the essay as stronger, thus s
uggesting that the difference between choice and no choice is that the inte
ntion is self-generated. Supporting this interpretation, no differences in
mail-back rates were observed in Studies 2 and 3 between a difficult and ea
sy choice condition. Furthermore, an alternative hypothesis that choice ind
uces implementation intentions did not receive support since independent ef
fects were in Study 2 found of deciding when and where to write the essay a
nd in Study 3 of making an outline of the essay.