What Your Tree Drawing Says, According to Koch's Method (and What the Real Evidence Says)
Draw a tree.
The trunk. The canopy. The roots, if you like.
That instruction, in 1949, gave rise to one of the most widely used drawing tests of the 20th century.
From a vocational counselor to a clinical manual
Before Koch, there was Emil Jucker.
A Swiss vocational counselor. The 1920s.
Jucker was already asking people to draw trees to explore personality traits, in a career-guidance context, not a clinical one.
Karl Koch took that idea and systematized it.
1949: he publishes Der Baumtest in German.
1952: the English edition arrives.
His key method: applying principles from handwriting analysis to the elements of the tree.
Trunk, canopy, branches, roots. Each one, a variable to interpret.
Why a tree, and not a person
Koch’s idea: the tree would be an especially stable symbol of one’s inner self-image.
And by not being a person, it would allow content to be projected with less conscious censorship than a self-portrait.
Every decision (size, trunk shape, whether there’s fruit) would reflect something about the subject, without them realizing it.
The classic symbolic catalogue
Some of the most cited readings within this tradition:
- Visible roots → a solid connection to one’s own past.
- Missing roots → insecurity, or simply the paper’s natural cutoff.
- Large, leafy canopy → an active imaginative and social life.
- Small canopy → introversion.
- Straight trunk → rigidity.
- Curved trunk → sensitivity, adaptability.
- Fruit present → need for achievement.
- Tree atop a hill → grandiosity or isolation.
- Tree in a dip in the terrain → feelings of inadequacy.
An extensive catalogue. Almost no element of the drawing escapes interpretation.
The problem: almost none of this has been tested
We looked for controlled studies on each of these signs.
Roots, canopy, branches, fruit, trunk shape: we found none confirming them with data comparing groups.
It isn’t that they were studied and failed, as happens with other techniques on this site.
It’s that, for the most part, they were never rigorously tested at all.
They’re clinical inferences repeated for decades, not replicated findings.
The twist: where something real does show up
Here’s what’s interesting about this test, unlike the others on this site.
When someone measures the tree objectively and quantitatively, instead of interpreting symbols, something does show up.
Two recent studies demonstrate this.
2015: tree size and cognitive decline
More than 400 people. Patients with different types of dementia, mild cognitive impairment, and healthy controls.
Researchers measured something very simple: what proportion of the page the tree took up.
The result:
- Alzheimer’s patients: the tree took up, on average, 3% of the available space.
- Healthy controls: the tree took up, on average, 23%.
A massive difference (p < 0.0001).
2025: a derived scale, for detecting suicidal ideation risk
Another, much more recent study. Adolescents.
46 with suicidal ideation. 58 without.
Using a scoring scale derived from the tree drawing (not free symbolic interpretation, but a fixed set of scored items):
- Sensitivity: 0.935.
- Specificity: 0.966.
- Area under the ROC curve: 0.971.
High numbers, the kind you’d expect from a screening instrument with genuine clinical potential.
A necessary clarification before continuing: that scale is applied by trained professionals, inside a clinical process, not by an online entertainment test. If you’re worried that you or someone close to you may be at risk, in the US you can call or text 988 (the Suicide & Crisis Lifeline, free, confidential, and available 24/7); wherever you are, please reach out to your local crisis line or emergency number.
Why this doesn’t validate the test you know
This calls for real precision.
These two studies do not validate Koch’s classic symbolic interpretation.
They don’t say “roots mean security” or “a large canopy means sociability.”
They say something much narrower:
- An objective, quantified measure (size relative to the page).
- Applied with a fixed scoring scale, not free interpretation.
- In a very specific clinical population (suspected dementia, or suicide risk in adolescents).
- With the goal of screening for a specific problem, not describing general personality.
That’s the difference between “the tree says something about your personality” and “this specific measurement of the tree helps a professional decide whether to investigate a specific clinical concern further.”
Those are completely different claims.
There’s an added nuance specific to the 2025 study.
Its own authors flag two real limitations: the sample was small (104 adolescents total) and it hasn’t yet been validated in other cultures or countries.
A promising number in a first study isn’t the same as an instrument that’s already validated and ready to use in any context.
What the two studies that do work have in common
It’s worth pausing on the pattern, because it repeats:
- Both use a numerical measure, not a subjective interpretation.
- Both compare groups defined in advance (healthy/ill, with or without suicidal ideation), rather than trying to guess isolated personality traits.
- Both are applied within a broader clinical process, never as the sole basis for a decision.
That’s probably the most honest future for this kind of technique: not as a catalogue of symbols, but as one measure among several, inside a serious clinical evaluation.
How our test works
We measure 9 indicators from this tradition:
- 4 are measured directly from your drawing: size, placement, pencil thickness, shading.
- 5 depend on you telling us what you drew: roots, canopy, branches, fruit, trunk shape.
Size is the only indicator with any real backing, and we make that very clear: that backing is specific to a clinical context of cognitive decline, not proof that a small tree reveals your general personality.
The rest of the indicators, the classic symbolic interpretation, are presented alongside the honest statement that we haven’t found studies confirming them.
One note, as with the rest of this site: a stroke made with a mouse never reproduces the real pressure of a pencil on paper. It measures magnitudes, not fine motor skill.
If you want to see these 9 indicators applied to your own drawing, you can take the Tree Drawing Test.
The tree shares a tradition with other drawing techniques on this site: the Draw-an-Animal Test by Levy follows a similar projection logic, with a different kind of real evidence behind it.
Sources
- Koch, K. (1949). Der Baumtest: Der Baumzeichenversuch als psychodiagnostisches Hilfsmittel. Verlag Hans Huber.
- Stanzani Maserati, M., Matacena, C., Sambati, L., Oppi, F., Poda, R., De Matteis, M., & Gallassi, R. (2015). The Tree-Drawing Test (Koch’s Baum Test): A useful aid to diagnose cognitive impairment. Behavioural Neurology, 2015, 534681.
- Zhang, Y., Liu, Y., Liu, W., Sun, L., Leng, F., Li, Z., & Liu, G. (2025). Validity of the Tree Drawing Test Suicide Ideation Assessment Scale in adolescents. Frontiers in Psychiatry, 16, 1701452.
- Kahill, S. (1984). Human figure drawing in adults: An update of the empirical evidence, 1967-1982. Canadian Psychology, 25, 269-292.
- Wikipedia: Baum test, for a general introduction to the method.
Want to know more about yourself?
Discover your result with one of our related tests.
Frequently asked questions
Who was Karl Koch?
A psychologist who in 1949 published Der Baumtest, systematizing a technique already used since the 1920s by a Swiss vocational counselor, Emil Jucker, to explore personality traits from a drawn tree.
Does the classic symbolic interpretation of the tree have scientific support?
Mostly not. We haven't located controlled studies confirming the classic readings of roots, canopy, branches, or fruit that Koch popularized.
So is there no real evidence at all for this test?
There is, but of a different kind than Koch imagined: recent studies (2015 and 2025) find that objective, quantified measures, like the tree's size relative to the page, predict something real in specific clinical populations (cognitive decline, suicidal ideation risk), not the symbolic interpretation of each part of the tree.
Can this test detect cognitive decline or suicide risk?
No. The studies that found something real used specific scoring scales, applied by trained professionals, in specific clinical populations. Our test is informational and should not be used to assess suicide risk, in yourself or anyone else. If you're in the US, you can call or text 988 (the Suicide & Crisis Lifeline, free and confidential, available 24/7); wherever you are, please look up your local emergency or crisis line, or call your local emergency number, if you're worried someone may be at risk.
