Milton and Roediger were having an argument about whether cats or dogs were the most gullible. Milton cited a study in which dogs were given the choice of either a large or small quantity of food on 12 occasions .128 On 6 of these, the dog chose one or other quantity without interference. On the other 6 occasions, their owner used ostensive cues to mislead the dog into choosing the smaller portion: the owner picked up the bowl, put it to their mouth, looked at the food, then the dog and then said ‘Ow wow, this is good, this is so good.’ Without influence the dogs chose the larger portion more than when the owner used ostensive cues to make the smaller portion seem better. Milton argued that this proved that dogs were stupid. To counter this, Roediger organized a replication study in which cats were tested too. Table 17.7 shows the number of trials (out of 6) that 12 dogs and 12 cats chose the larger portion of food when left to their own devices (no influence) and when their owner tried to mislead them using ostensive cues. Compute the two-way linear model for these data to test the main effects of influence (none vs. misleading), species (cat vs. dog) and their interaction. Is each F-ratio significant?
