WEBVTT

00:00:06.545 --> 00:00:08.985
In a study in the 1990s,

00:00:08.985 --> 00:00:13.887
participants recalled getting lost 
in a shopping mall as children.

00:00:13.887 --> 00:00:17.207
Some shared these memories
in vivid detail—

00:00:17.207 --> 00:00:20.887
one even remembered that the old man
who rescued him

00:00:20.887 --> 00:00:23.863
was wearing a flannel shirt.

NOTE Paragraph

00:00:23.863 --> 00:00:28.148
But none of these people 
had actually gotten lost in a mall.

00:00:28.148 --> 00:00:30.378
They produced these false memories

00:00:30.378 --> 00:00:35.096
when the psychologists conducting 
the study told them they’d gotten lost,

00:00:35.096 --> 00:00:37.696
and although they might not remember
the incident,

00:00:37.696 --> 00:00:40.456
their parents had confirmed it.

00:00:40.456 --> 00:00:45.311
And it wasn’t just one or two people 
who thought they remembered getting lost—

00:00:45.311 --> 00:00:48.161
a quarter of the participants did.

NOTE Paragraph

00:00:48.161 --> 00:00:50.921
These findings may sound unbelievable,

00:00:50.921 --> 00:00:54.571
but they actually reflect 
a very common experience.

00:00:54.571 --> 00:00:57.821
Our memories are sometimes unreliable.

00:00:57.821 --> 00:01:01.821
And though we still don’t know precisely
what causes this fallibility

00:01:01.821 --> 00:01:03.591
on a neurological level,

00:01:03.591 --> 00:01:07.591
research has highlighted some 
of the most common ways our memories

00:01:07.591 --> 00:01:10.591
diverge from what actually happened.

NOTE Paragraph

00:01:10.591 --> 00:01:14.431
The mall study highlights how we can
incorporate information

00:01:14.431 --> 00:01:16.161
from outside sources,

00:01:16.161 --> 00:01:18.391
like other people or the news,

00:01:18.391 --> 00:01:22.566
into our personal recollections
without realizing it.

00:01:22.566 --> 00:01:27.129
This kind of suggestibility is just 
one influence on our memories.

00:01:27.129 --> 00:01:28.249
Take another study,

00:01:28.249 --> 00:01:32.574
in which researchers briefly showed 
a random collection of photographs

00:01:32.574 --> 00:01:34.174
to a group of participants,

00:01:34.174 --> 00:01:39.817
including images of a university campus
none of them had ever visited.

00:01:39.817 --> 00:01:42.367
When shown the images three weeks later,

00:01:42.367 --> 00:01:47.246
a majority of participants said 
that they had probably or definitely

00:01:47.246 --> 00:01:50.076
visited the campus in the past.

NOTE Paragraph

00:01:50.076 --> 00:01:55.669
The participants misattributed information
from one context— an image they’d seen—

00:01:55.669 --> 00:02:01.067
onto another— a memory of something
they believed they actually experienced.

00:02:01.067 --> 00:02:05.545
In another experiment, people were shown
an image of a magnifying glass,

00:02:05.545 --> 00:02:09.065
and then told to imagine a lollipop.

00:02:09.065 --> 00:02:13.994
They frequently recalled that they saw 
the magnifying glass and the lollipop.

00:02:13.994 --> 00:02:17.464
They struggled to link the objects 
to the correct context—

00:02:17.464 --> 00:02:21.464
whether they actually saw them,
or simply imagined them.

NOTE Paragraph

00:02:21.464 --> 00:02:25.464
Another study, where a psychologist 
questioned over 2,000 people

00:02:25.464 --> 00:02:29.044
on their views about the legalization 
of marijuana,

00:02:29.044 --> 00:02:32.494
highlights yet another kind 
of influence on memory.

00:02:32.494 --> 00:02:38.462
Participants answered questions 
in 1973 and 1982.

00:02:38.462 --> 00:02:43.361
Those who said they had supported
marijuana legalization in 1973,

00:02:43.361 --> 00:02:46.841
but reported they were against it in 1982,

00:02:46.841 --> 00:02:53.402
were more likely to recall that they were 
actually against legalization in 1973—

00:02:53.402 --> 00:02:57.790
bringing their old views in line 
with their current ones.

00:02:57.790 --> 00:03:00.920
Our current opinions,
feelings, and experiences

00:03:00.920 --> 00:03:04.920
can bias our memories 
of how we felt in the past.

00:03:04.920 --> 00:03:06.060
In another study,

00:03:06.060 --> 00:03:10.262
researchers gave two groups
of participants background information

00:03:10.262 --> 00:03:16.672
on a historical war and asked them to rate
the likelihood that each side would win.

00:03:16.672 --> 00:03:19.232
They gave each group the same information,

00:03:19.232 --> 00:03:23.539
except that they only told one group
who had actually won the war—

00:03:23.539 --> 00:03:27.139
the other group didn’t know 
the real world outcome.

00:03:27.139 --> 00:03:30.549
In theory, both groups’ answers 
should be similar,

00:03:30.549 --> 00:03:32.839
because the likelihood
of each side winning

00:03:32.839 --> 00:03:35.519
isn’t effected by who actually won—

00:03:35.519 --> 00:03:39.879
if there’s a 20% chance of thunderstorms,
and a thunderstorm happens,

00:03:39.879 --> 00:03:44.976
the chance of thunderstorms
doesn’t retroactively go up to 100%.

00:03:44.976 --> 00:03:48.366
Still, the group that knew 
how the war ended

00:03:48.366 --> 00:03:53.956
rated the winning side as more likely 
to win than the group who did not.

NOTE Paragraph

00:03:53.956 --> 00:03:58.195
All of these fallibilities of memory 
can have real-world impacts.

00:03:58.195 --> 00:04:03.502
If police interrogations use leading
questions with eye witnesses or suspects,

00:04:03.502 --> 00:04:10.813
suggestibility could result in incorrect 
identifications or unreliable confessions.

00:04:10.813 --> 00:04:13.393
Even in the absence of leading questions,

00:04:13.393 --> 00:04:17.775
misattribution can lead to inaccurate 
eyewitness testimony.

00:04:17.775 --> 00:04:18.655
In a courtroom,

00:04:18.655 --> 00:04:21.685
if a judge rules a piece of evidence 
inadmissible

00:04:21.685 --> 00:04:26.263
and tells jurors to disregard it,
they may not be able to do so.

00:04:26.263 --> 00:04:29.773
In a medical setting, if a patient 
seeks a second opinion

00:04:29.773 --> 00:04:33.773
and the second physician is aware 
of the first one’s diagnosis,

00:04:33.773 --> 00:04:37.403
that knowledge may bias their conclusion.

00:04:37.403 --> 00:04:41.403
Our memories are not ironclad 
representations of reality,

00:04:41.403 --> 00:04:43.753
but subjective perceptions.

00:04:43.753 --> 00:04:46.653
And there’s not necessarily
anything wrong with that—

00:04:46.653 --> 00:04:50.653
the problems arise when we treat
memory as fact,

00:04:50.653 --> 00:04:53.023
rather than accepting
this fundamental truth

00:04:53.023 --> 00:04:56.063
about the nature of our recollections.