Remember: The Science of Memory and the Art of Forgetting

Neuroscientist and novelist Lisa Genova, author of Still Alice, draws on her dual expertise to explain how human memory works, why it fails, and what people can do to improve it. She opens by asking readers to picture a penny from memory, then challenges them to recall details such as which direction Lincoln faces or where the word LIBERTY appears. Most people cannot, despite thousands of exposures. This inability, she argues, is not a memory failure but memory functioning as designed: The brain remembers what is meaningful and forgets what is not.

Genova identifies four steps of memory creation: encoding (translating perceptions into neurological language), consolidation (linking neural activity into an associated pattern), storage (maintaining that pattern through structural changes), and retrieval (reactivating the connections). Central to this process is the hippocampus, a seahorse-shaped brain structure that binds disparate sensory information into a retrievable unit. She illustrates the hippocampus's importance through the case of Henry Molaison, who in 1953 had both hippocampi surgically removed to treat severe epilepsy; the seizures largely subsided, but Molaison lost the ability to form new long-term memories for the remaining 55 years of his life. Fully consolidated memories, Genova explains, do not reside in the hippocampus but are distributed throughout the brain, and retrieval is reconstruction rather than playback.

Genova argues that attention is the essential prerequisite for memory formation: If a person does not pay attention to something, no memory is created. She illustrates this with a personal anecdote about losing her car in a parking garage, not because her memory failed but because she never registered where she parked. She extends the principle to forgetting names, noting that a name spoken during an introduction exists in the brain for only 15 to 30 seconds; without deliberate attention, it vanishes. The brain's default state is inattentive, and multitasking significantly decreases the likelihood of forming memories.

She then introduces working memory, the limited holding space in the prefrontal cortex for whatever one is paying attention to right now. Citing psychologist George Miller's 1956 finding, she explains that working memory holds roughly seven items (plus or minus two) for 15 to 30 seconds. Information can be sustained through repetition or compressed through chunking, such as grouping a 10-digit phone number into three segments. Working memory serves as the gateway to long-term storage: Details carrying special meaning or emotion can be sent to the hippocampus for consolidation.

Genova identifies three types of long-term memory. Muscle memory, or procedural memory, stores physical skills. Consolidated by the basal ganglia, a motor-control brain structure, rather than the hippocampus, these memories are retrieved unconsciously and remain stable over decades, explaining why Molaison could learn new motor tasks despite having no conscious memory of practicing them. Semantic memory stores facts and knowledge one "just knows" without recalling the circumstances of learning. Genova stresses that self-testing doubles retention compared with rereading and that meaning dramatically strengthens recall, citing studies of taxi drivers and chess masters whose advantages vanished when information lacked meaningful context. Episodic memory records personal experiences tied to a specific place and time. Routine events are quickly discarded, while emotion, surprise, and meaning activate the amygdala, an emotion-processing brain structure that signals the hippocampus to consolidate the experience. Most autobiographical memories cluster between ages 15 and 30, a phenomenon called the reminiscence bump, likely because those years contain the most meaningful firsts.

The book's second part turns to why memories fail. Genova argues that episodic memories are inherently inaccurate, vulnerable to distortion at every processing stage. Each time a memory is recalled, it becomes susceptible to editing and is reconsolidated with changes that overwrite the original. She cites studies in which 25 to 50 percent of subjects insisted they remembered fabricated events, and describes the classic car-crash experiment by psychologists Elizabeth Loftus and John Palmer, in which substituting the word "smashed" for "contacted" caused subjects to estimate cars traveling 10 miles per hour faster. She connects these findings to the criminal justice system, noting that approximately 75 percent of 365 DNA exonerations in the United States as of 2019 involved convictions based on eyewitness testimony.

Genova examines tip-of-the-tongue states, the common experience of knowing a word but being unable to produce it. She explains the "ugly sister" phenomenon, in which a loosely related decoy word blocks retrieval of the target. The Baker/baker paradox illustrates a related principle: The word "baker" as an occupation connects to rich associations, while "Baker" as a surname is a neurological dead end, explaining why people remember details about someone but not their name. Genova also addresses prospective memory, the brain's system for remembering future intentions, which is so unreliable she characterizes it as almost a form of forgetting. She prescribes practical aids: to-do lists, calendar alerts, pillboxes, and placing cues in impossible-to-miss locations.

Time, Genova argues, is the primary enemy of stored memories. She presents memory researcher Hermann Ebbinghaus's 1885 forgetting curve, which showed rapid initial memory loss leveling off over time, and identifies two defenses: repetition and meaning. Genova reframes forgetting as sometimes adaptive, citing the case of Solomon Shereshevsky. Studied by Russian psychologist Alexander Luria for 30 years, Shereshevsky could memorize enormous volumes of information but found his inability to discard irrelevant data a profound daily burden. Forgetting outdated information, Genova argues, clears mental space for what matters.

Genova distinguishes normal age-related memory decline from Alzheimer's disease. Normal aging brings increased tip-of-the-tongue frequency, slower processing speed, and reduced working memory and episodic recall, while muscle memory and semantic knowledge remain stable or grow. Alzheimer's, by contrast, is driven by the progressive accumulation of amyloid beta protein, which forms harmful plaques in brain synapses. The disease begins in the hippocampus and spreads to regions governing spatial orientation, executive function, emotion, and eventually motor skills. Genova draws on her friendship with Greg O'Brien, an acclaimed journalist diagnosed with early-onset Alzheimer's at 59, who describes forgetting conversations from minutes earlier and standing in front of his yellow Jeep without recognizing he had driven there. She stresses the distinction: Misplacing keys and finding them on the table is normal; finding them in the refrigerator, or not knowing what keys are for, signals Alzheimer's.

The final section addresses factors that enhance or impair memory. Context-dependent memory means retrieval improves when conditions match those during encoding. Acute stress enhances memory formation related to the stressor but impairs retrieval of already-stored memories, while chronic stress shrinks the hippocampus, inhibits neurogenesis (the growth of new neurons), and increases Alzheimer's risk. Sleep is essential: It replays and consolidates new memories, boosts recall by 20 to 40 percent compared to equivalent waking time, and clears amyloid from the brain during deep sleep. Adults need seven to nine hours nightly. Genova outlines prevention strategies: The Mediterranean or MIND diet, which emphasizes leafy greens, berries, nuts, whole grains, and fish, reduces Alzheimer's risk by a third to a half; aerobic exercise, even a daily brisk walk, correlates with 40 percent reduced risk; and learning new things builds cognitive reserve. Cognitive reserve refers to the redundancy of neural connections that buffers against Alzheimer's pathology, as demonstrated by the Nun Study, in which some participants showed full Alzheimer's pathology at autopsy yet displayed no symptoms during life.

Genova closes by proposing a balanced relationship with memory: taking it seriously enough to care for it while holding it lightly enough to forgive its imperfections. She returns to Greg, whose humor, faith, and capacity for connection persist despite devastating memory loss, and to her grandmother, who no longer recognized her family yet still knew she was loved. Memory, Genova concludes, is not everything. A person is more than what they can remember.