The Google effect, also called digital amnesia, is the finding that when people expect information to remain available, they remember where to find it rather than the thing itself. It comes from Betsy Sparrow and colleagues in 2011, it is one of the most-cited psychology findings of the internet era, and it has a replication problem that is almost never mentioned alongside it.
Definition#
The Google effect: a shift in what people encode to memory when they believe information will remain externally accessible, favouring the location or retrieval route over the content. It describes a change in what is remembered rather than a general decline in the ability to remember.
The original finding#
Sparrow, Liu and Wegner published a set of experiments in Science in 2011. Participants told that statements they typed would be saved recalled the statements less well, and recalled the folder they were saved in better, than participants told the statements would be erased. The framing drew on Wegner's earlier concept of transactive memory, the idea that couples and teams routinely divide remembering between them and treat other people as memory stores.
That lineage matters. Externalising memory into other people is ancient and generally works well. The 2011 work extended it to machines rather than discovering something new about human frailty.
The part that is usually left out#
Subsequent replication attempts have produced mixed results, and the effect has been part of the wider replication difficulties in social psychology from that period. Some studies have found it, some have not, and effect sizes have generally been smaller than the original.
This does not mean the phenomenon is imaginary. It means the confident version, that search engines are measurably degrading human memory, rests on considerably weaker ground than its popularity implies. It is quoted as an established fact roughly a thousand times more often than its evidential status warrants, and a research site that used it that way would be doing the thing this research criticises elsewhere.
The sturdier adjacent evidence is behavioural rather than about memory as such. Habitual satnav users showed worse unaided spatial memory, with steeper decline over three years of heavier use, which is a longitudinal design in a specific domain rather than a laboratory manipulation of expectation.
How it differs from cognitive offloading#
They are related and routinely conflated. Cognitive offloading is the broader and better-evidenced concept: using an external tool or physical action to reduce the mental demand of a task, defined and measured by Risko and Gilbert in 2016. The Google effect is one specific instance of it, concerning memory encoding under an expectation of availability.
If you want the concept with the stronger evidence base, use offloading. If you specifically mean the memory-for-location finding, use the Google effect and cite it accurately.
Does it apply to AI?#
Plausibly, and nobody has established it. The mechanism is different in an important way: a search engine returns a source you still have to read and judge, while a generative system returns a conclusion. Offloading the location of information and offloading the reasoning about it are not the same act, and the second has no equivalent of the 2011 experiment behind it.
The Google effect is a useful analogy for what generative AI might do to memory, and an analogy is not evidence. This research lists durable cognitive effects of long-term AI use as unknown, and that includes this.
When the trade is reasonable#
Remembering where to find something is a reasonable allocation rather than an automatic loss. It is what people have always done with libraries, colleagues and notebooks. The question worth asking is narrower: can you still evaluate the thing you retrieved?
You cannot verify an answer in a domain where you never built competence. That is why the practical case for knowing things rests on knowledge being the instrument you check machine output with, rather than on nostalgia about memorisation. Offload the storage; keep the judgement that lets you tell whether what came back is right.
Related SuperSkills research#
On the broader and better-evidenced concept, cognitive offloading. On thinking effects, AI and critical thinking. On what is and is not established, what we actually know. On practice, using AI without dependency. See should AI remember everything about me.
Key sources
- Sparrow, B., Liu, J. and Wegner, D. M. (2011). Google effects on memory. Science, 333(6043).
- Risko, E. F. and Gilbert, S. J. (2016). Cognitive offloading. Trends in Cognitive Sciences, 20(9).
- Dahmani, L. and Bohbot, V. D. (2020). Habitual use of GPS negatively impacts spatial memory. Scientific Reports, 10.
About this definition#
The Google effect belongs to Sparrow, Liu and Wegner, not to SuperSkills. It appears here with its replication difficulties stated, because it is one of the most over-claimed findings in this field and this research would rather lose the rhetorical convenience than repeat it uncritically. Rahim Hirji is the author of SuperSkills (Kogan Page, 2026).
About this research#
Written by Rahim Hirji, author of SuperSkills (Kogan Page, 2026), keynote speaker on AI and human capability, and founder of The SuperSkills Intelligence Company.
Cite this
Hirji, R. (2026). What is the Google effect? The SuperSkills Intelligence Company. Last reviewed 26 August 2026. thesuperskills.com/research/what-is-the-google-effect