Research demonstrates that young children’s reports are more likely to be influenced by misleading information than are the reports of older children or adults.

This could reflect either

(1) Young children’s memories are more subject to distortion
(2) Children’s are simply responding to social influence and their memories per se are not distorted.

New issues in the present study

(1) Delay: In the Cassel study the children answered the recognition questions shortly after answering the miseading questions. This may have magnified the consistency pressures on the children.  Afterall they had just said "yes" to the leading question, it would be pretty inconsistent not to select the suggested item on the recognition test.
(2) Interviewer: In the Cassel study the misleading questions and the recognition questions were asked by the same interviewer. Chlidren may have felt obliged to go along with the interviewer both on the misleading questions and on the recognition memory test.

They then completed the first interview

Free Recall
Unbiased Questions: “Can you tell me who owned the bike?”
Misleading Questions: “The girl was wearing a dress, wasn’t she?”

Free Recall
Three alternative forced choice (e.g. “Was the girl wearing (a) long dark jeans (b) a bathing suit or (c) a dress?”).

Free Recall: Free recall was very accurate for all age groups, with older age groups recalling more information than younger age groups.

Unbiased Questions: Participants could respond in one of three ways (1) accurately (2) inaccurately (3) “I don’t know”.

(1) Correct Responses increased with age
(2) Incorrect Responses decreased with age

(1) Correct responses (i.e. “no”) increased with age
(2) Incorrect responses (i.e. “yes”) decreased with age [affirmation bias?]

Free Recall: Accurate free recall decreased with age. Again little false recall, but there was somewhat more false recall on day 2 than day 1 and for younger children than for older children.

Three Alternative Force Choice

Neutral Items:

(1) Interviewer had little effect. One exception is for 7 year olds there were more correct responses in the same interviewer condition.
(2) In addition correct responses seemd to increase with age.

(1) Interviewer had little effect on target selection. Exception was for adults where they were more likely to answer correctly when the same interviewer was used.
(2) Interviewer did have a significant effect on false recognition of suggested item  However it was in the opposite of what was predicted.  The Same Interviewer condition resulted in lower rates of false recognition of selected items.

They predicted that using the same interviewer across both sessions would increase social demand pressures resulting in more errant responses on the critical questions (i.e. questions they had been misled about) on day 2.

The opposite appeared to happen if anything. Adults were less accurate on critical questions following a two day delay

Bjorklund et. al argue that this result is a function of reinstatement of context effects.

They restate their argument for the social demand characteristics account. That is, kids and adults who assent to a leading question don’t always stick with that on later recognition tests or free recall tests.