2. The Sin of Absent-mindedness

The Sin of Absent-mindedness

On a brutally cold day in February 1999, seventeen people gathered

in the nineteenth-floor office of a Manhattan skyscraper to compete for a
title known to few others outside that room: National Memory Champion.
The winner of the U.S. competition would go on to the world memory
championship several months later in London.

The participants were asked to memorize thousands of numbers and
words, pages of faces and names, lengthy poems, and rearranged decks of
cards. The victor in this battle of mnemonic virtuosos, a twenty-seven-
year-old administrative assistant named Tatiana Cooley, relied on classic elaborative encoding techniques: generating visual images, stories, and associations that link incoming information to what she already knows. Given
her proven ability to commit to memory vast amounts of information, one
might also expect that Cooley's life would be free from the kinds of mem-
ory problems that plague others. Yet the champ considers herself dangerously forgetful. "I'm incredibly absent-minded," Cooley told a reporter.

Fearful that she will forget to carry out everyday tasks, Cooley depends on
to-do lists and notes scribbled on sticky pads. "I live by Post-its," she admitted ruefully.
The image of a National Memory Champion dependent on Post-its
has a paradoxical, even surreal quality: Why does someone with a capacity
for prodigious recall need to write down anything at all? Can't Tatiana
Cooley call on the same abilities and strategies that she uses to memorize
hundreds of words or thousands of numbers to help remember that she
needs to pick up a jug of milk at the store? Apparently not: the gulf that
separates Cooley's championship memory from her forgetful everyday life
illustrates the distinction between transience and absent-mindedness.

The mnemonic techniques that Cooley has mastered help her to
counter the effects of the sin of transience. Give ordinary people a long
string of numbers to memorize, and by the time they have gone much past

the seventh or eighth digit, the first few on the list have faded. Not so for a
skilled mnemonist like Cooley, who has encoded the numbers in a manner
that makes them readily accessible even when time passes and more numbers are encoded. But the kinds of everyday memory failures that Cooley
seeks to remedy with Post-it Notes - errands to run, appointments to
keep. and the like - have little to do with transience. These kinds of memory failures instead reflect the sin of absent-mindedness: lapses of attention
that result in failing to remember information that was either never encoded properly (if at all) or is available in memory but is overlooked at the
time we need to retrieve it.

To appreciate the distinction between transience and absent-mindedness, consider the following three examples:

A man tees up a golf ball and hits it straight down the fairway. After
waiting a few moments for his partner to hit, the man tees up his
ball again, having forgotten that he hit the first drive.
A man puts his glasses down on the edge of a couch. Several minutes
later, he realizes he can't find the glasses, and spends a half-hour
searching his home before locating them.
A man temporarily places a violin on the top of his car. Forgetting
that he has done so, he drives off with the violin still perched on
the roof.
Superficially, all three examples appear to reflect a similar type of
rapid forgetting. To the contrary, it is likely that each occurred for very different reasons.

The first incident took place back in the early 1980s, when I played golf
with a patient who had been taking part in memory research conducted in
my laboratory. The patient was in the early stages of Alzheimer's disease,
and he had severe difficulties remembering recent events. Immediately after hitting his tee shot, the patient was excited because he had knocked it
straight down the middle; he realized he would now have an easy approach
shot to the green. In other words, he had encoded this event in a relatively
elaborate manner that would ordinarily yield excellent memory. But when
he started teeing up again and I asked him about his first shot, he expressed
no recollection of it whatsoever. This patient was victimized by transience:
he was incapable of retaining the information he had encoded elaboratively, and no amount of cueing or prodding could bring it forth.


In the second incident, involving misplaced glasses, entirely different
processes are at work. Sad to say, this example comes from my own experience and happens more often that I would care to admit. Without at-
tending to what I was doing, I placed my glasses in a spot where I usually do
not put them. Because I hadn't fully encoded this action to begin with
my mind was preoccupied with a scientific article I had been reading.

I was at a loss when I realized that my glasses were missing. When I finally
found them on the couch. I had no particular recollection of having put
them there. But unlike the problem facing the golfing Alzheimer's patient,
transience was not the culprit: I had never adequately encoded the information about where I put my glasses and so had no chance to retrieve it later.
The third example, featuring the misplaced violin. turned into far
more than just a momentary frustration. In August 1967, David Margetts
played second violin in the Roth String Quartet at UCLA. He had been entrusted with the care of a vintage Stradivarius that belonged to the Department of Music. After Margetts put the violin on his car roof and drove
off, UCLA made massive efforts to recover the instrument. Nonetheless,
it went missing for twenty-seven years before resurfacing in 1994, when
the Stradivarius was brought in for repair and a dealer recognized the instrument. After a lengthy court battle, the violin was returned to UCLA in
1998.

There is, of course, no way to know exactly what Margetts was thinking about when he put the violin on the roof. Perhaps he was preoccupied
with other things. just as I was when I misplaced my glasses. But because
one probably does not set down a priceless Stradivarius without attending
carefully to one's actions, I suspect that had Margetts been reminded before
driving off, he would have remembered perfectly well where he had just
placed the violin. In other words, Margetts was probably not sabotaged by
transience, or even by failure to encode the event initially. Rather, forgetting
in Margett's case was likely attributable to an absent-minded failure to notice the violin at the moment he needed to recall where he had put it. He
missed a retrieval cue - the violin on the car roof - which surely would
have reminded him that he needed to remove the instrument.

Absent-minded memory failures are both amusing and frightening.
To understand the basis for them, we need to probe the role of attention in
encoding processes, and also to explore how retrieval cues and reminders
help us remember what we intend to do.


We have already seen that the degree and type of elaborative encoding that
people carry out can strongly affect transience. When such encoding fails
altogether, however, conditions are ripe for the annoying kinds of absentminded memory failures that sometimes seem a regular part of daily existence: misplaced glasses, lost keys, forgotten appointments, and so forth.
One way to prevent elaborative encoding is to disrupt or divide attention
when people are acquiring new information. In studies of divided attention, experimental participants are given a set of target materials to remember, such as a list of words, a story, or a series of pictures. At the same
time, they are required to perform an additional task that draws their attention away from the to-be-remembered material. For example, people might
be asked to monitor an ongoing series of tones and to respond when they
hear a high-pitched or low-pitched tone, while at the same time they try to
study a list of words for a later test. Or, while studying the words, they
might be asked to listen to a series of numbers and respond whenever a series of three consecutive odd numbers appears. Compared to a condition
in which they are allowed to pay full attention to the study list, people exhibit extremely poor memory for the words studied under divided attention conditions.

Recent studies suggest that dividing attention during encoding does
not necessarily prevent people from registering some information about an
experience. Memory researchers have found it useful to distinguish between two ways in which we remember past experiences: recollection and
familiarity. Recollection involves calling to mind specific details of past experiences, such as exactly where you sat in the restaurant you dined at last
week, the tone of voice used by the waiter who served you, or the kind of
spices used in the Cajun-style entree that you ordered. Familiarity entails a
more primitive sense of knowing that something has happened previously,
without dredging up particular details. In the restaurant, for example, you
might have noticed at a nearby table someone you are certain you have met
previously despite failing to recall such specifics as the person's name or
how you know her. Laboratory studies indicate that dividing attention during encoding has a drastic effect on subsequent recollection, and has little
or no effect on familiarity. This phenomenon probably happens because
divided attention prevents us from elaborating on the particulars that are
necessary for subsequent recollection, but allows us to record some rudimentary information that later gives rise to a sense of familiarity. When attention is divided, we may still record enough information about a face so
that it seems familiar when we encounter it again, even though we do not
engage in sufficient elaboration to recollect the person's name, occupation,
or other details later.

Many absent-minded errors are probably attributable to a kind of "divided attention" that pervades our daily lives. Mentally consumed with
planning for a critical presentation the next day, you place your car keys in
an unusual spot as you are reading over your notes. Or, thinking about how
much money is left in your checking account after writing the latest check,
you leave the checkbook on the dining room table. Even if some residual
familiarity remains from these encounters, it is not sufficient to prevent
forgetting later on: you need to be able to recollect the details of where you
put the keys or the checkbook. Lew Lieberman, a sixty-seven-year-old retired psychology professor, relates a particularly irritating incident of this
kind:

A day does not go by when I do not spend time looking for something.
Today I needed a new booklet of checks for my checkbook. When I
went to get it, I found the very next booklet was missing. Apparently, at
some earlier time, I could not find my checkbook and had to use a
check from the next booklet to write a check, but then I could not find
the missing booklet and have NO recollection of having done this. But
then, where is the booklet?
Insufficient attention at the time of encoding may be an especially important contributor to absent-minded errors in older adults. A series of experiments carried out by the psychologists Fergus Craik and Larry Jacoby
indicates that aging can produce a state that resembles a kind of chronic divided attention. They found similar patterns of memory performance in
older adults (aged sixties to seventies) who are allowed to pay full attention
to incoming information during encoding and college students whose attention is divided during encoding. For instance, in Jacoby's experiments
both groups showed less recollection of past experiences than did college
students who paid full attention at the time of encoding, even though all
three groups showed similar levels of familiarity. Dividing attention reduces the overall amount of cognitive resources - the "energy supply"
that fuels encoding - that can be devoted to incoming information. Likewise, Craik and others argue that aging is associated with a decline in cognitive resources, thereby resulting in patterns of performance that resemble
those produced by divided attention.

Attention collapses that yield absent-minded forgetting are particularly
likely for routine activities that do not require elaborative encoding. During the early stages of performing complex activities, such as driving a car
or typing, we need to pay careful attention to every component of the ac-
tivity. But as skill improves with practice, less and less attention is required
to perform the same tasks that initially demanded painstaking effort. Numerous experiments have shown that practice on various kinds of tasks
and skills results in a shift from attention-demanding, effortful task execution to automatic execution involving little or no deployment of attention.
"Operating on automatic" provides us with the cognitive freedom to focus
on unrelated matters as we perform what once was an attention-consuming task, such as driving a car. But automaticity has a cost: the virtual absence of recollection for activities that were performed "on automatic:'

Most seasoned drivers, for example, are familiar with the unsettling experience of cruising along at sixty-five miles per hour on a six-lane interstate,
and suddenly realizing that they have no recollection of the road for the
past five miles. Absorbed with concerns that have nothing to do with driving, and relying on the well-learned skills that allow them to drive safely
even when on automatic, the experienced driver does not elaborate on
what is going on around him and hence remembers nothing of it. Over a
century ago, the British novelist Samuel Butler, who developed a grand theory of mental evolution that assigned great importance to the development
of automatic behavior, insightfully characterized memory for automatic
actions in a concert pianist who has just played a five-minute piece:

For of the thousands of acts ... which he has done during the five min utes, he will remember hardly one when it is over. If he calls to mind
anything beyond the main fact that he has played such and such a
piece, it will probably be some passage which he has found more difficult than the others, and with the like of which he has not been so
long familiar. All the rest he will forget as completely as the breath
which he has drawn while playing.

This kind of amnesia for the automatic can lead to some jarring incidents of forgetting. It is probably responsible for the kind of forgetting I experienced when, on automatic, I put my eyeglasses down in an unlikely lo-
cation. Even worse, people report frantically searching for glasses that, only
moments ago, they casually pushed up on top of their heads, or running
around the house looking for keys they are still holding. My own most frustrating "amnesia for the automatic" story occurred after finishing a round
of golf last summer. I carried my clubs back to my car and prepared for the
drive home. I usually put my car keys in my golf bag during the round, but
could not find the keys there. Panicking, I emptied the contents of the bag
to no avail. I couldn't find the keys in my pockets and, assuming they had
fallen out of the bag when I was playing, began silently cursing under my
breath as I contemplated what to do next. Out of the corner of my eye, I
then noticed the raised trunk of the car with the keys dangling from the
back. Operating on automatic, I had already used the keys to open my
trunk but had no memory of it.

Neuroimaging techniques are starting to provide insights into what
happens in the brain during conditions of divided attention and automatic
behavior. Tim Shall ice and his collaborators performed PET scans while
volunteers tried to learn a list of word pairs. Some scans were conducted
while people performed an easy, distracting task that diverted little attention away from encoding the word pairs: volunteers moved a bar in the
same predictable direction on all trials. Other scans were carried out while
the volunteers performed a difficult, distracting task that drew most of
their attention away from encoding the word pairs: they moved the bar in
novel, unpredictable directions on each trial. There was less activity in the
lower left part of the frontal lobe during the difficult distraction scans than
during the easy distraction scans. As we saw in the previous chapter, activation in the lower left frontal region during encoding is closely related to
subsequent remembering and forgetting. Shallice's experiment suggests
that dividing attention prevents the lower left frontal lobe from playing its
normal role in elaborative encoding. When this region is not involved in
encoding new information, or only minimally involved, subsequent recollection will suffer greatly, and absent-minded types of forgetting are likely
to occur.

Related neuroimaging studies also link the left inferior frontal lobe
with automatic behavior. The neuroscientist Marcus Raichle and his group
performed PET scans while they showed volunteers a series of common
nouns and asked them to generate related verbs. So, for example, when
shown the noun dog, participants might come up with bark or walk. When
subjects first performed this task, generating verbs was associated with ex-
tensive activity in the lower left frontal lobe (and many other parts of the
brain). This activity probably reflects a kind of elaborative encoding related
to thinking about properties of dogs, and the kinds of actions they per-
form. But, as the volunteers practiced the task repeatedly with the same
nouns, and generated the verbs more quickly and automatically, activity
in the lower left frontal lobe gradually declined. This result raises the possibility that automatic behaviors in everyday life - a key source of absent-
minded errors - may be associated with low levels of left prefrontal activity.

In a more recent fMRI study conducted by Anthony Wagner in my
laboratory, we saw further evidence of how automatic behavior, reflected
by reduced activity in the left inferior prefrontal cortex, works against
forming vivid recollections. Memory researchers have known since the pioneering studies of Herman Ebbinghaus over a century ago that repeating
information improves memory for what is repeated. Further, distributing
the repetitions over time often results in better memory than massing them
all together. So, for instance, if you want to study for a test you will be taking in a week's time, and are able to go through the material ten times, it is
better to space out the ten repetitions during the week than to squeeze
them all together (students often engage in massive cramming just before
taking an exam, which can produce short-term gains in retention, but
spacing out repetitions generally produces better long-term results).

We showed people words to encode for a later test, either a day before
we showed them the same words again in the scanner (spaced repetition),
or just a few minutes before they saw the same words again (massed repetition). Predictably, people showed better memory on the test for the spaced
words than for the massed words. Most important, there was less activity
in the left inferior prefrontal region when people studied the massed words
they had just seen a few minutes earlier than when they studied the spaced
words they had seen a day earlier. Repeating the words close together in
time apparently led to more automatic encoding on the second repetition,
which was associated with reduced left prefrontal activity and poorer subsequent memory. These results fit nicely with those from Raichle's verbgeneration experiment, and might help us to understand why automatic
kinds of encoding can lead to absent-minded memory errors.

Automatic or superficial levels of encoding can result in other kinds of
absent-minded errors, too. One of the most intriguing is known as "change
blindness." In studies of change blindness, people observe objects or scenes
that unfold over time. Experimenters make subtle or large changes to the
objects or scenes in order to determine whether people notice the changes.
Change blindness occurs when people fail to detect the changes that the experimenter has made. The psychologists Daniel Levin and Daniel Simons
have performed some of the most inventive research on change blindness.
In one study, for instance, they showed participants a movie in which a
young blond man sits at a desk. He then gets up, walks away from the desk,
and exits the room. The scene then shifts outside the room, where the
young man makes a phone calL Unknown to the observers, the man sitting
at the desk is not the same person as the man who makes the phone call (although both are young, blond, and wear glasses, they are clearly different
people when examined at all carefully). Only one-third of observers no-
ticed the change.

In another film, two women are shown sitting across a table from each
other, sipping colas and munching on food as they chat. As the camera cuts
back and forth between the two, it all seems pretty normal and mundane.
When asked if they notice whether anything changed during the brief duration of the film, most people say they did not detect any changes, or per-
haps noticed one. Yet in every frame there were numerous changes in the
women's clothes, props on the table, and so forth.

Not satisfied with merely demonstrating change blindness in film segments, Levin and Simons asked whether such effects could also be demon-
strated in live interactions. To test this idea, one experimenter asked someone on a college campus for directions. While they were talking, two men
walked between them carrying a door that hid a second experimenter. Behind the door, the two experimenters traded places, so that when the men
carrying the door moved on, a different person from the one who had been
there just a second or two earlier was now asking for directions. Remarkably, only seven of fifteen participants reported noticing this change!

In successive experiments, Simons has demonstrated even more dramatic effects by further restricting attention to an object. Consider this scenario: if you were watching a circle of people passing a basketball, and
someone dressed in a gorilla costume walked through the circle, stopped to
beat his chest, and exited, of course you would notice him immediately
wouldn't you? Simons and the psychologist Chris Chabris filmed such a
scene and showed it to people who were asked to track the movement of
the ball by counting the number of passes made by one of the team. Approximately half of the participants failed to notice the gorilla.

Focused on tracking the ball's movement, people are blind to what
happens to unattended objects and thus do not encode the sudden change.
Brain imaging evidence from a related experimental procedure supports
this idea. When people are instructed to pay attention to letter strings superimposed on line drawings of objects, parts of the left frontal, temporal,
and parietal lobes respond more strongly to meaningful words than to ran-
dom letters. But when they are instructed to pay attention to the line drawings, these regions no longer respond differently to words and random letters - even though participants look directly at the letter strings.

In the earlier examples of change blindness, where people are free to
attend to whatever they wish, change blindness probably occurs because
people encode features of a scene at an extremely shallow level, recording
the general gist of the scene but few of the specific details. To paraphrase
Simons and his collaborators, successful change detection tends to occur
when people encode elaboratively the exact features that distinguish the
original object or person from the changed one. In the "door study," people
who most often failed to notice that a different person emerged from behind the door were middle-aged and older adults; college students tended
to notice the change. Older individuals might have encoded the initial
(young) experimenter generically as a "college student," whereas the college
students (for whom the person asking directions was a peer) encoded the
experimenter in a more specific way. To find out whether college students
would be more susceptible to change blindness when induced to encode at
a generic level, Simons and Levin repeated the "door study" attired as construction workers. College students now might tend to encode them more
generically and, hence, show higher levels of change blindness. And they
did: only four of twelve students noticed when a different construction
worker emerged from behind the door to ask instructions. Thus, shallow
encoding that does not proceed beyond a general level results in poor recollection of the details of a scene and consequent vulnerability to change
blindness. Change blindness is attributable, at least in part, to the same
kinds of automatic encoding activities that sometimes leave us searching
for glasses perched atop our foreheads or keys clenched in our fists.

REMEMBERING WHAT You WANT TO Do.

In Marcel Proust's monumental exploration of his own memory, Remem-
brance of Things Past, the author's yearning to recapture lost moments from
childhood seems to epitomize what memory is for: providing a connection
between the present and the past. Yet in our daily lives, memory is just as
much about the future as it is about the past. We are all familiar with the
seemingly endless to-do lists that remind us of what we need to remember
in the future. Pick up milk and cereal on the way home; call to make that
airline reservation; drop off a manuscript at an associate's office; confirm
tomorrow's lunch date; mail in the mortgage payment on time; transfer
money from savings to checking - the list could go on indefinitely.

Psychologists nowadays use the term "prospective memory" to describe remembering to do things in the future. Until recently, researchers
had focused almost exclusively on the remembrance of things past which
constituted the object of Proust's yearnings and writings, even though people express more concern about remembering to carry out future actions
than about other, retrospectively oriented aspects of memory. This distinction may be because when retrospective remembering fails - forgetting a
name or a fact, or confusing when and where two events occurred-
memory is seen as unreliable. But when prospective remembering fails -
forgetting a lunch appointment or failing to drop off a package as promised
- the person is seen as unreliable. Have you ever forgotten to send in your
monthly mortgage check or credit card payment? If so, you know that
faulty memory is not a sufficient excuse to escape the late-payment fine.
Absent-minded errors of prospective memory are annoying not only be-
cause of their pragmatic consequences, but also because others tend to see
them as reflecting on credibility and even character in a way that poor retrospective memory does not.

Why does prospective memory fail? To begin to answer this question,
I find it useful to adopt a distinction first proposed by the psychologists
Gilles Einstein and Mark McDaniel. They distinguish between "event-
based" and "time-based" prospective memory. Event-based prospective
memory involves remembering to carry out a task when a specific event oc-
curs. If your friend Frank says, "When you see Harry at the office today, tell
him to call me;' Frank is asking you to remember to perform a particular
action (tell Harry to call me) when a specific event occurs (you see Harry at
the office). Time-based prospective memory, in contrast, involves remembering to carry out an action at a specific time in the future. Remembering
to take the cookies out of the oven in twenty minutes or remembering to
take your medicine at 11:00 P.M. are examples of time-based prospective
memory tasks.

Forgetting can occur for different reasons when we are faced with
event-based and time-based prospective memory tasks. In event-based
tasks, problems occur if the event that is designated to trigger recall of the
intended action fails to do so: if, for instance, we see Harry in the office and
are not reminded to tell him to call Frank. In time-based tasks, by contrast,
problems usually arise because we fail to encounter or generate a cue that
can remind us to carry out the target action. When faced with the task of
remembering to take medicine at 11:00 P.M., either I must spontaneously
remember to do so at 11:00 P.M., or think ahead and arrange cues that will
likely trigger recall at the right time. Knowing that I am likely to be brushing my teeth before bed at 11:00 P. M., for example, I might place the medicine in a spot by the sink where I can't miss it. From this perspective, event-
based prospective memory requires understanding of why cues or hints do
or do not spontaneously trigger recall of an intended action; time-based
prospective memory requires understanding of how we generate cues that
will help us to remember at a later time.

Consider first event-based prospective memory. Frank has asked you
to tell Harry to call him, but you have forgotten to do so. You indeed saw
Harry in the office, but instead of remembering Frank's message you were
reminded of the bet you and Harry made concerning last night's college
basketball championship, gloating for several minutes over your victory
before settling down to work. When Frank later asks you what happened
with Harry, you apologize profusely and wonder out loud whether something has gone terribly wrong with your memory. In all likelihood, nothing
is wrong. Prospective memory failed because Harry is a potential reminder
of many things other than the message to call Frank. The best prospective
memory triggers tend to be highly distinctive cues that have few other associations in long-term memory, and hence are not likely to remind us of irrelevant information.

Experiments by Gilles Einstein and Mark McDaniel using a simple
laboratory analogue of event-based prospective memory demonstrate the
importance of cue distinctiveness. Participants were given lists of words to
learn for a later test. For the prospective memory task, some people tried to
remember to push a button whenever a particular familiar word appeared,
such as movie, and others were instructed to push the button whenever a
particular unfamiliar item appeared, such as the nonsense word yolif Einstein and McDaniel reasoned that people have many associations to movie
and, hence, might sometimes think of them instead of remembering to
press the button. They have no associations to yolif and, hence, would not
be distracted by irrelevant information and forget to push the button. Re-
sults indeed showed that people were much more likely to remember to
press the button when cued with yolif than with movie.
A reminder also has to be sufficiently informative, as well as distinctive, to aid prospective recall. How many times have you jotted down a
phone number that you need to call, thinking that it will remind you later
to do so, only to discover that you do not recollect whose phone number it
is? When I visited a college campus to lecture on memory, my host's secretary showed me a "reminder" she had scrawled on a sticky pad earlier that
day which said only "Nat." She now had no idea who or what she meant by
"Nat." When we write a note to ourselves, all the surrounding information
is available in working memory, so the reminder seems perfectly adequate.
But we may fail to take into account the main lesson in the previous chap-
ter: memories are frequently transient. The reminder that seemed self-evident when related information was available in working memory becomes
a cryptic - and frustrating - puzzle when that information has faded
with time. To aid future recall, we need to transfer as many details as possible from working memory to written reminders.
Event-based prospective memory can also fail because we are preoccupied with other concerns and devote so little attention to the target event
that we are not spontaneously reminded of anything. If you saw Harry in
the office only minutes before you had to give a major presentation to your
CEO, you may have been devoting so much mental effort to preparing for
the talk that the sight of Harry failed to trigger any recollections at all. Experiments using a variant of the Einstein-McDaniel procedure support this
possibility. People were shown a series of words and were instructed to remember to press a button whenever a particular word appeared. Some participants also performed an additional attention-demanding task. For example, in one experiment some participants rapidly tried to generate a
sequence of digits in a random order while they were also studying the
word list and remembering to perform the prospective memory task.

Compared to participants who were allowed to generate digit sequences at
a more leisurely pace, the group that generated digits rapidly showed many
more lapses of prospective memory - that is, they forgot to press the but-
ton when the target word appeared. They were preoccupied with trying to
generate random digit sequences quickly, so the word cues frequently failed
to remind them of what they were supposed to do, much as someone
preoccupied with preparing a talk might fail to pass on a message to a
coworker she encounters while in the midst of preparations. In other experiments in which participants were given relatively mindless additional
tasks to perform, such as continually repeating the word the while studying
words and trying to remember to press a button when the target appeared,
prospective memory did not suffer.

A recent study that used PET scans to examine brain activity during
an event-based prospective memory task further illuminates these findings.
While in the scanner, participants were instructed to repeat a series of spoken words. In a condition that also required prospective memory, they
tried to remember to tap whenever a designated target word was spoken.
Compared to when participants repeated words but did not have to re-
member to carry out a future action, prospective remembering was associated with greater activity in several parts of the frontal lobe. Some of these
same frontal regions have been implicated previously in working memory
- holding information on-line for brief time periods. Although we do not
yet know how these laboratory findings relate to everyday absent-minded
errors, it is tempting to speculate that some of the frontal regions that
showed heightened activity during prospective remembering are "captured" by distracting activities that preoccupy us and contribute to failed
prospective memory. Consider, for example, what might happen when we
are told to pass on a message to an associate, but are preoccupied with
competing task demands - thinking about what we said, or didn't say, at
this morning's meeting when we encounter that associate. Some of the
frontal regions that contribute to successful prospective remembering may
be tied up by our internal monologue, and thus do not play their usual role
in enabling prospective recall. This could result in a failure to be reminded
by a cue to carry out an intended action.

Stressed-out baby boomers who worry that each new absent-minded
memory slip signals the onset of age-related cognitive decline, or perhaps
even Alzheimer's disease, should take comfort in the finding that prospective cues frequently fail to trigger recall of appropriate actions when people
are preoccupied with attention-consuming matters. The source of the worried boomers' difficulties may well lie in the multitude of competing professional and personal concerns that absorb mental energy and can reduce
the effectiveness of reminders to carry out mundane but necessary everyday tasks. Indeed, several laboratory studies have shown that older adults
who have reached their sixties and seventies perform almost as well as
younger adults on event-based prospective memory tasks. When given cues
that remind them to carry out a target task, older adults have little problem
remembering what to do.

Aging does have more noticeable effects on time-based prospective
memory tasks. When we need to carry out an action at a particular time in
the future, such as remembering to take medication before bed, we must
generate cues or reminders on our own. For example, in laboratory studies
by Einstein and McDaniel, older and younger adults were instructed to remember to press a key after ten and twenty minutes had passed; a clock was
positioned behind the subjects to help them keep track of time. In this
time-based task, older adults forgot to press the key more often than did
younger. With no cue available to trigger recall of the target action, older
adults were less likely than younger adults to summon up the action on
their own. This finding fits well with other data indicating that self-initiated recall is a difficult task for older adults, probably because it requires
extensive cognitive resources that decline with age.

Older adults can, however, perform well on time-based prospective
memory tasks by taking steps to convert them into event-based tasks, that
is, by generating cues that will be available at the appropriate time to trig-
ger recall of what must be done. When asked by an experimenter to make a
phone call at a specified time, some older adults changed the time-based
task to an event-based task by linking it with an incident in their daily lives
that occurred when the call had to be made. For example, one participant
placed a reminder note to make the call next to where she washed dishes
and another tied in the phone call with a morning coffee.

These findings have implications for such important everyday prospective memory tasks as taking medications. Many older adults need multiple medications, and taking them at the right time is crucial for their
health. Surveys suggest that between one-third and one-half of elderly
adults do not adhere to their medication schedules. Direct observations indicate that such problems are mainly characteristic of people in their seventies and eighties; "young" old adults (in their sixties) generally adhere
well to medication schedules. As noted earlier, remembering to take medication at 11:00 P.M. is a time-based task, but it can be converted to an
event-based task by, say, placing the needed medications next to one's
toothbrush, and regularly brushing one's teeth before retiring at 1l:00 P.M.
Many factors contribute to poor medication adherence, but improvements
can be realized by arranging cues that convert this time-based task to an
event-based task.

Perhaps even more than event-based prospective memory, time-based
prospective memory often fails because people are preoccupied with other
concerns that prevent them from even attempting to generate appropriate
retrieval cues. In the study that required participants to make a phone call
at a particular time, the most common reasons they gave for failing to do so
were that they were "absorbed" or "distracted." Unfortunately, we often
merely admonish ourselves to remember to carry out a task at a future
time, rather than generating concrete cues or reminders that will help to do
so. Sitting at a desk in your home office, you tell yourself earnestly, "OK,
don't forget to mail that credit card payment tomorrow morning." But unless you convert this time-based task to an event-based task by generating a
reminder, such as putting the bill in a place where you will see it when you
leave for work the next morning, the bill will likely remain unsent on top of
your desk. The psychologist Susan Whitbourne related to me a particularly
vexing incident of this sort:

In leaving for an overnight trip to Baltimore, I "told" myself to be sure
to pack my contact lens case in the morning, after I was through with it
at home. However, I forgot to do so, as I found when I looked for it in
my bag that night. Spotting two empty water glasses with handy paper
covers, I thought I would put one lens in each glass and cover them up
and that would do it. At the time, I was quite weary after a long day of
travel and an evening of socializing. In the morning, I went to the sink
and, to my horror, saw that the right glass had been removed and was
empty! The glass of water I helped myself to in the middle of the night
had an extra little treat in it that would never make it to my eye.
Fortunately, I was able to give my talk that day wearing only one contact lens,
but it was a pretty miserable experience, not to mention an expensive
memory slip.

Despite the health risk of swallowing a contact lens, Whitbourne's absent minded error led to a relatively benign, if irritating, outcome. But in
other contexts, more serious consequences can follow from failed time-
based prospective recall. Air traffic control provides a compelling example.
Controllers are frequently faced with situations in which they must postpone an action and remember to carry it out at a later time. For example,
if a pilot requests a higher altitude that cannot be granted until nearby
aircraft pass, the controller must remember to give the clearance later. To
help remember, controllers make use of rectangular strips of paper, called
"flight progress strips;' which provide information about the altitude,
route, destination, and other features of each flight for which the controller
is responsible. A controller who defers a request for higher altitude, for example, might try to use the strip for that flight as a reminder by marking it
or offsetting it from other strips.

Flight progress strips will be replaced eventually by automated electronic strips that do not allow controllers to manipulate them physically. To
help determine how controllers can use such reminders most effectively, re-
searchers at the University of Oklahoma collaborated with the Federal Aviation Administration in a simulated study of air traffic control. Consider a
controller who has just deferred a request from Delta flight 692 for higher
altitude until passing traffic clears, and enters a prospective command to
remind himself to grant higher altitude to Delta 692 in one minute. One
possibility would be to make the electronic reminder visible during the
minute waiting period to help the controller "rehearse" the command, but
not at the moment the command is to be executed. Another possibility
would be to make the electronic reminder visible only at the moment when
the command needs to be retrieved and executed. Yet a third possibility
would be to keep the reminder visible both during rehearsal and at the mo-
ment of intended retrieval. Compared to a condition in which no electronic reminder was provided, prospective recall improved only when the
cue was available at the time needed for retrieval. Providing the reminder
during rehearsal alone produced no benefit, and providing the reminder
during both rehearsal and at the moment of retrieval was no more effective
than providing it at retrieval alone.

The importance of having a cue available at the time an intended action is carried out, rather than beforehand, was painfully illustrated when I
received a call at home one morning from my wife. She reminded me to

leave cash for our housecleaners, who would make their weekly visit later
that day. She also reminded me not to set the security alarm, because the
cleaners do not know the code. I immediately removed the cash needed to
pay them and placed it on the kitchen table. I then resumed what I had
been doing (writing this very chapter) and left for the office later that
morning. Two hours later, I received a message from a friend who had been
notified by our security company that the blaring siren alarm in our house
had gone off. Police arrived quickly, and the cleaners faced the awkward
task of explaining that they intended to clean - not to clean out - our
house. My wife's reminder to leave money worked because I was able to act
on it immediately. Her reminder to refrain from setting the alarm induced
me to "rehearse" - admonishing myself not to forget, just as Susan Whit-
bourne told herself to pack the case for her contact lenses. But the reminder
ultimately failed because it was not present when I needed to avoid setting
the alarm at the time I left the house a couple of hours later.

Because prospective memory so heavily depends on the availability of
cues that trigger intended actions, the most effective way to counter absentminded prospective memory failures is to develop and use effective external memory aids. The most effective such cues pass two key criteria considered earlier: they are sufficiently informative, and are available at the time
an action needs performing. The quintessential external memory aid - a
string tied around one's finger - passes the latter of these two criteria, but
not the former. Tying a string around a finger is potentially helpful because
it's always visible. But it renders us vulnerable to the same kind of problem
that frustrated the secretary who puzzled over what she meant by the reminder saying only "Nat.": it is easy to forget what a string around the
finger means. Even if we do write down sufficiently detailed notes to remind us of what we intend to do, we still must find a way to ensure that
they are available around the time the action is to be performed. Sticky-pad
notes hidden in our pockets or a notebook that we never look at may contain all the necessary details but will not solve problems unless they are
consulted.

A number of elementary and secondary schools have instituted effective programs that use external memory aids to combat a common absent minded error among students: forgetting to do homework. For example, in
one Atlanta-area elementary school, students record assignments to be carried out in a central planning notebook that parents are asked to sign each
night. The school's principal spot-checks the planners, awarding ice cream
and candy kisses to those students whose parents sign off every day of the
week. To encourage its use, at one high school the planner serves as a hall
pass, and at another students are required to carry it to the water fountain
or restroom. Informal reports suggest a reduction of forgotten homework
assignments.

Many effective everyday memory aids that we take for granted meet
the two key criteria of informativeness and availability at the time of retrieval. A whistling teakettle, for instance, reminds you of exactly what you
need to do at the time you need to do it. Likewise, some electronic irons
come with a built-in prospective memory cue, sounding an alarm when left
face-down for too long. Even more sophisticated electronic devices are now
available to help us record and plan our future actions. A survey conducted
in the early 1990S identified thirty different kinds of external memory aids
that were then available commercially, and the list has no doubt grown
longer during the past decade. Interestingly, there were variations in the
perceived usefulness of different types of external aids as a function of age
and life style. Young adults in their teens and twenties tended to be most interested in "high-tech" reminders such as electronic memo pads that can be
used at school or on the job. Middle-aged adults with families viewed
products that reduce forgetting of tasks around the house, such as the
"memory iron," as particularly helpful. And older, mainly retired adults
were most interested in products that help to execute routine daily tasks at
home and elsewhere, such as an electronic "plant reminder" that is inserted
into soil and sounds when the plant needs to be watered.

When Joseph Tsien and his group published their groundbreaking
study of genetically engineered memory improvements in mice, the media
were awash in speculations about high-tech memory drugs that might put
an end to forgetting altogether. But just as 1999 National Memory Champion Tatiana Cooley still forgets to do things and struggles to overcome her
absent-minded memory failures, there is likewise no guarantee that any future drugs that combat transience would also reduce absent-mindedness.
As Cooley discovered, however, combating absent-mindedness does not require genetic interventions: the Post-its that she relies on, or other more sophisticated external memory aids, are adequate remedies when used effectively. Absent-mindedness is most troublesome for busy individuals who
are perpetually trying to balance multiple tasks and must therefore con-


stantly organize future actions. The psychologist Ellen Langer has pointed
out that when we misplace our car keys or eyeglasses, it is usually because
we are devoting our mental resources to more important things: wrestling
with a personal dilemma or thinking about how to handle an upcoming
meeting at work. Are there also absent-minded mice, preoccupied with
pressing concerns that lead to automatic behaviors and associated forgetting? Might there be a specific gene responsible that could help to over-
come such memory failures? If one exists, would we want to make use of it?
These are intriguing questions without clear answers. But I suspect that for
the foreseeable future, cognitive engineers, not genetic engineers, will lead
the way forward in efforts to combat absent-mindedness.