NBEO Part I common mistakes
NBEO Part I · Understand the traps and how to avoid them.
When a question shows a lesion or pathway and asks for the resulting visual field defect, the candidate answers from the anatomical name alone (for example, "optic tract lesion" → "homonymous hemianopia") without working out which retinal fibers are interrupted and therefore which part of the field is lost, and without checking whether the defect is congruent or incongruent or whether it respects the vertical or horizontal meridian.
Candidates memorize lesion-to-defect labels as paired facts rather than deriving the defect from the crossing and non-crossing fiber arrangement. Under time pressure the label is retrieved faster than the reasoning, and the label is often correct only for the classic complete lesion, not for the partial or post-chiasmal variants the exam favors.
What to do instead
For every lesion question, draw the pathway and label nasal versus temporal fibers at each segment. State the rule explicitly: nasal fibers cross and carry the temporal field; temporal fibers stay ipsilateral and carry the nasal field. Then locate the lesion relative to the crossing and decide which eye loses which half-field. Practice with partial lesions (incongruent tract versus congruent cortex, Meyer loop versus parietal radiation, chiasmal compression versus junctional lesion with the ipsilateral temporal hemianopia and contralateral upper temporal loss). Confirm laterality by naming the side of the brain and the side of the field before selecting an answer.
In optics problems the candidate substitutes numbers into a formula without first fixing a sign convention and a coordinate direction, so that object distance, image distance, focal length, and vergence terms are assigned signs inconsistently within the same problem, producing a plausible but wrong answer.
Many candidates learn several sign systems (real-is-positive, Cartesian, vergence) and switch between them mid-calculation, especially when a problem mixes mirrors, thin lenses, and reduced eyes. The arithmetic feels correct because each individual step is defensible, so the error is not caught by re-reading the question.
What to do instead
Adopt one convention for the whole exam and write it at the top of scratch paper: light travels left to right, distances measured against the incident light are negative, converging power positive. Convert every problem to vergence form (L = n/l, L′ = L + F, l′ = n′/L′) so that signs are handled by the algebra rather than by intuition. After computing, sanity-check the sign against the physical situation (a real object in front of a converging lens must give a real image behind it when the object is outside the focal point) and check the magnification sign against whether the image should be inverted.
When a case presents with a systemic disease, the candidate jumps to the named ocular association from memory (for example, "rheumatoid arthritis → scleritis") without using the specific details given (laterality, chronicity, associated lab findings, medication history) to discriminate among the several ocular conditions that disease can cause.
Review resources present systemic disease as one-line associations, so candidates store a single headline link per disease. Exam items deliberately supply discriminating features that point to a less famous manifestation or to a drug side effect rather than the headline association.
What to do instead
For each major systemic disease in the syllabus, build a short differential list of its ocular manifestations rather than one association, and attach the discriminating feature to each (for example, which manifestation is bilateral, which is sight-threatening, which correlates with a specific antibody, which is caused by the treatment rather than the disease). When reading an item, underline the discriminating details first and only then recall the list. Train yourself to ask whether the finding is explained by the disease or by its therapy.
In binocular vision and motility items, the candidate answers from the deviation's name or from the cover test result alone, without determining whether the deviation is comitant or incomitant, which eye is fixating, and how the deviation changes in different gazes and at near versus distance.
Cover test interpretation is taught early and becomes automatic, so candidates stop at "esotropia" or "exotropia" and skip the additional measurements the question is really testing. Incomitance and fixation patterns are the discriminating features, and they are easy to omit when the answer choices include the simple label.
What to do instead
Use a fixed sequence for every motility item: establish fixation preference, measure deviation at near and distance with and without correction, compare versions in all gazes, and test for a pattern (A or V) and for oblique muscle involvement. State whether the deviation is comitant or incomitant and what that implies (refractive, accommodative, paretic, restrictive, or mechanical). Only then match to the answer. Practice interpreting a described Parks three-step result and a described prism alternate cover test in the same sitting so the sequence becomes habitual.
On biostatistics and epidemiology items, the candidate computes a sensitivity, specificity, predictive value, or odds ratio correctly but then answers the question that was asked about the wrong quantity, or reports a measure without checking whether the study design and sampling permit it.
Candidates drill the formulas but not the distinctions among them. Predictive values depend on prevalence while sensitivity and specificity do not, and relative risk is estimable from cohort but not from case-control sampling. Items are written to exploit exactly these distinctions, and a correct calculation can still yield a wrong choice.
What to do instead
Before computing, write down what the question asks for in words (for example, "probability of disease given a positive test" versus "probability of a positive test given disease"). Build a two-by-two table for every item and fill cells from the stem rather than manipulating formulas in the abstract. Memorize the design constraints: prevalence-dependent versus prevalence-independent measures, which effect measure each design supports, and what confidence intervals and p-values do and do not tell you. Check whether the answer you computed is even available from the design described.
In pharmacology items the candidate selects a drug by its class name or by the condition it treats, ignoring the receptor selectivity, the route, and the adverse-effect profile that the stem's details point to, and ignoring interactions with the patient's other medications or systemic conditions.
Drug lists are memorized as condition-to-drug pairs, which is sufficient for straightforward items but fails when the question tests why one agent in a class is preferred, or which agent is contraindicated given a comorbidity. Candidates also under-weight the pharmacokinetic and adverse-effect details that are the actual discriminator.
What to do instead
For each drug class, organize knowledge by mechanism, receptor selectivity, route, onset and duration, and characteristic adverse effects, and note the specific clinical situation that favors or forbids each agent (for example, which agents are risky in asthma, which in narrow angles, which in pregnancy). Read the stem for comorbidities and concurrent medications before looking at the options, and eliminate options on contraindication grounds first. Rehearse the autonomic receptor table until selectivity is automatic.
On microbiology and immunology items the candidate answers from the organism's name or Gram stain alone, without using the described host (contact lens wearer, immunocompromised, neonate, post-surgical) and the described clinical pattern (time course, lesion morphology, culture or stain findings) to identify the likely organism or immune mechanism.
Organism lists are memorized by category, and candidates treat "bacterial keratitis" as one entity. The exam distinguishes organisms by host risk factor and by clinical tempo, which are precisely the details candidates skim.
What to do instead
Build organism profiles keyed to host and tempo: which organisms dominate contact lens-associated keratitis, which follow trauma with soil contamination, which occur in the immunocompromised, which present with a specific stain or culture characteristic. For immunology, separate the four hypersensitivity mechanisms and attach a canonical ocular example to each, plus the meaning of a positive skin test, an anergy panel, and a vaccine response. Read the stem's host description first and use it to narrow the organism list before reading options.
On visual perception and psychophysics items the candidate confuses the stimulus parameter with the perceptual response, or confuses the measurement method with the threshold it estimates, and answers from the phenomenon's name rather than from the mechanism the question describes.
Terminology in this domain is dense and visually similar (adaptation versus sensitization, increment threshold versus detection threshold, spatial frequency versus contrast), and candidates recognize the term without being able to state what varies and what is measured. Exam items often describe a procedure and ask what it measures, which recognition alone cannot answer.
What to do instead
For each phenomenon, write a one-line description of the independent variable, the dependent variable, and the mechanism. Practice converting a described experiment into that triple before reading options. Separate the psychophysical methods (method of adjustment, limits, constant stimuli, forced choice) from the thresholds and functions they estimate, and separate retinal from cortical contributions to adaptation, contrast sensitivity, and acuity. Use a small set of worked examples to anchor each distinction.
On professional, ethical, and legal items the candidate answers from personal values or general ethics intuition rather than from the specific obligation the scenario triggers, and does not distinguish among confidentiality, informed consent, duty to warn or report, scope of practice, and record-keeping duties.
These items read like common sense, so candidates answer quickly without identifying which duty is at issue. The legally correct action often differs from the intuitively kind or loyal action, and the exam tests that gap.
What to do instead
Classify each scenario before answering: is the question about consent capacity, a mandatory report, a confidentiality exception, a scope-of-practice boundary, or documentation? For each category, know the governing principle and the usual correct action (for example, when a duty to warn or report overrides confidentiality, and what must be documented and to whom). Practice by writing the duty at issue in one sentence before reading options, and eliminate any option that resolves the scenario by violating a duty rather than by following the applicable one.
The candidate manages time and item difficulty poorly: spending excessive time on a small number of calculation-heavy optics or biostatistics items, then rushing the remaining items, and changing answers on review without a specific reason.
Calculation items feel solvable with more effort, so candidates persist past the point of diminishing return, and the resulting time pressure degrades performance on items they would otherwise answer correctly. Answer-changing is driven by anxiety rather than by a newly identified error.
What to do instead
Set a per-item time budget from the total item count and section timing, and enforce a rule to mark and move on after a fixed interval on any calculation item. Return only to marked items with remaining time. On review, change an answer only when you can name the specific error in the original reasoning; otherwise keep the first response. Simulate the timing under exam conditions with full-length practice sets so the budget becomes automatic.