Approach-avoidance motivation
Also known as: Approach and avoidance motivation
Behavior is energized either by moving toward what's good or by moving away from what's bad — and the two systems differ.
What it means
Approach-avoidance motivation is the foundational distinction between behavior directed by positive, desired stimuli (approach) and behavior directed by negative, undesired stimuli (avoidance). Rooted in a valence asymmetry that recurs across psychology and neuroscience — often mapped onto separate behavioral activation and inhibition systems — the dimension organizes phenomena from goal pursuit to personality to emotion. Approach motivation is linked to eagerness, exploration, and positive affect; avoidance motivation to vigilance, caution, and negative affect, and it tends to be more potent because threats demand priority. The nuance is that a single goal can be framed either way ('get an A' versus 'don't fail'), with avoidance framing generally costlier for well-being. It matters as a unifying scaffold beneath regulatory focus, achievement goals, and temperament.
Why the two systems differ
The difference is not just which way you walk. Elliot's hierarchical model stacks the layers: a temperament on top, a specific goal in the middle, outcomes at the bottom, so one underlying sensitivity can be channelled into very different goals depending on how a situation is read. The two directions also do different computational work. Approach motivation defines success as a state to reach: an open set of routes count, and arrival is something you can verify. Avoidance motivation defines success as the absence of a bad outcome, a standard you can never confirm you have met, since no run of nothing-going-wrong proves nothing will. The only strategy left is to keep scanning for signs of failure. That asymmetry in confirmability, not valence alone, is why Elliot argues the avoidance route carries a regulatory cost even when it produces identical behavior.
What the evidence shows
Elliot and Church's 1997 classroom study is the template: mastery goals grew out of achievement motivation and high competence expectancies, performance-avoidance goals out of fear of failure and low expectancies, and only performance-avoidance goals damaged both intrinsic motivation and grades. Van Yperen, Blaga and Postmes's 2014 meta-analysis paired self-reported goals with independently measured performance across 98 papers and 33,983 people, and found small but consistent effects in the expected direction — mastery-approach r = .14, performance-approach r = .10, performance-avoidance r = -.13 — but domain moderated them: the avoidance penalty appeared in education (r = -.14) and work (r = -.20) and vanished in sport (r = -.04, ns). The causal claim rests on a separate meta-analysis by the same team in 2015, covering 19 experiments that induced a goal state rather than measuring it, where approach goals did outperform avoidance goals and mastery-approach beat performance-approach — but particularly when people anticipated no feedback and faced no time pressure. That is narrower than the framework's popular version. The antecedent side is weaker still: Lochbaum and colleagues' 2017 meta-analysis of the sport and physical education literature put most links between the model's proposed antecedents and goal adoption at roughly r = .10 to .33, small enough that the authors doubted their practical significance.
Limits and caveats
The neural story is the weakest part. Greater left-frontal cortical activity has long been read as an approach marker, but Kuper, Kaeckenmester and Wacker's 2019 meta-analysis of 79 samples found resting frontal asymmetry explained under 0.4% of the variance in extraversion and neuroticism, too little to serve as a trait index. The behavioral measures have their own trouble: Loijen and colleagues' 2020 review found that while approach-bias modification has some purchase in addiction, the equivalent training for anxiety produced no clinical effects despite an avoidance bias that shows up reliably in the lab, and it flagged recurring method problems across this literature — missing control groups and control stimuli, and samples too small to power the comparisons being drawn. Valence and direction also come apart, since anger is negatively valenced but approach-motivated. Treat the dimension as an organizing scaffold, not a mechanism you can measure off someone's scalp.
Using it in practice
The lever you actually hold is framing, and it costs nothing to pull. Default to approach language when you want exploration, persistence and enjoyment: name the state to be reached rather than the one to be dodged. Reserve avoidance framing for cases where it is honest and the stakes are asymmetric, such as safety procedures, compliance, and error-critical work, where vigilance is the point and narrowing is a feature. Watch for accidental avoidance framing in metrics and dashboards, which get built out of failure counts and so quietly install an avoidance goal in everyone who reads them. And the evidence on dispositional antecedents is weak enough that you should not read someone's temperament off a single framing choice.
Examples
Studying to earn distinction (approach) feels and functions differently from studying to escape the shame of failing (avoidance), even when the grade is identical.
A sales team chasing a bonus tries odd new leads and unusual pitches; the same team told not to miss quota plays it safe with accounts it already holds.
'Eat five vegetables a day' and 'cut out the junk' aim at the same dinner, but the first sends you shopping with curiosity and the second turns every meal into a test you can fail.
A checklist item reading 'confirm flaps set' names a state you can verify and tick; 'don't forget the flaps' names a failure you can only keep scanning for, never confirm.
Anger feels bad but moves you toward its target: the insulted person leans in, argues back and closes the distance, while fear at the same intensity backs away. Bad feeling and backing off are separate axes.
First described in Roots in Lewin and Atkinson; modern synthesis by Andrew Elliot (1999, 2006).
Key references
- Loijen, A., Vrijsen, J. N., Egger, J. I. M., Becker, E. S., & Rinck, M. (2020). Biased approach-avoidance tendencies in psychopathology: A systematic review of their assessment and modification. Clinical Psychology Review, 77, 101825. doi.org/10.1016/j.cpr.2020.101825
- Kuper, N., Kaeckenmester, W., & Wacker, J. (2019). Resting frontal EEG asymmetry and personality traits: A meta-analysis. European Journal of Personality, 33(2), 154-175. doi.org/10.1002/per.2197
- Lochbaum, M., Jean-Noel, J., Pinar, C., & Gilson, T. (2017). A meta-analytic review of Elliot's (1999) Hierarchical Model of Approach and Avoidance Motivation in the sport, physical activity, and physical education literature. Journal of Sport and Health Science, 6(1), 68-80. doi.org/10.1016/j.jshs.2015.07.008
- Van Yperen, N. W., Blaga, M., & Postmes, T. (2014). A meta-analysis of self-reported achievement goals and nonself-report performance across three achievement domains (work, sports, and education). PLoS ONE, 9(4), e93594. doi.org/10.1371/journal.pone.0093594
- Van Yperen, N. W., Blaga, M., & Postmes, T. (2015). A meta-analysis of the impact of situationally induced achievement goals on task performance. Human Performance, 28(2), 165-182. doi.org/10.1080/08959285.2015.1006772
- Elliot, A. J. (2006). The hierarchical model of approach-avoidance motivation. Motivation and Emotion, 30(2), 111-116. doi.org/10.1007/s11031-006-9028-7
- Elliot, A. J., & Thrash, T. M. (2002). Approach-avoidance motivation in personality: Approach and avoidance temperaments and goals. Journal of Personality and Social Psychology, 82(5), 804-818. pubmed.ncbi.nlm.nih.gov/12003479/