📝 Limits and continuity of vector functions (25 MCQs)

📖 Explanation: This question requires evaluating the limit of each component independently as $t \to 0$. The first component approaches 0. The second requires recognizing the standard limit form where $\sin(u)/u \to 1$ as $u \to 0$, yielding 0 since the numerator vanishes faster than the denominator in this specific configuration? No, actually $\sin(t^3)/t \approx t^3/t = t^2 \to 0$. Wait, re-evaluating: if it were $\sin(t)/t$, it would be 1. Here $\sin(t^3)/t \to 0$. The third component $e^{-1/t^2} \to 0$. Thus the correct vector is $\langle 0, 0, 0 \rangle$. However, option B is a common distractor for students who misapply the sine limit rule without checking powers. The correct answer is actually A based on rigorous calculation, making B the realistic misconception distractor. Students must verify each component's asymptotic behavior rather than assuming standard limits apply universally.

📖 Explanation: Determining continuity requires finding the intersection of domains for all scalar components. The square root requires $t \geq 1$. The natural log requires $2-t > 0$ or $t < 2$. The rational term excludes $t=3$. Intersecting $[1, \infty)$, $(-\infty, 2)$, and $\mathbb{R} \setminus \{3\}$ yields $[1, 2)$. Note that $t=3$ is already excluded by the log constraint. Option A incorrectly includes values where the log is undefined. Option B misses the valid endpoint at $t=1$ where the square root is continuous from the right within the domain. This tests precise set intersection skills over simple computation.

📖 Explanation: While $\|\mathbf{r}(t)\| \to 0$ implies $\mathbf{r}(t) \to \mathbf{0}$ by the squeeze theorem, a non-zero norm limit does not guarantee vector convergence. For example, a vector could rotate indefinitely while maintaining constant length, meaning the direction never settles. Therefore, knowing only the magnitude converges is insufficient to prove the vector limit exists unless that magnitude is zero. Option C captures this nuance for general $L$, but D is the only statement that is *necessarily* true across all cases including $L=0$. This distinguishes between necessary conditions and sufficient conditions in vector analysis.

📖 Explanation: Oscillation alone doesn't preclude limits (e.g., damped oscillations). However, here $\cos(1/t)$ and $\sin(1/t)$ do not settle to unique values, so the vector limit indeed fails to exist. The student's conclusion is right, but their reasoning is imprecise. Discontinuity arises because the projection onto the xy-plane traverses the unit circle infinitely often without converging to a point, violating the uniqueness requirement of limits. Simply stating 'oscillation' is vague; one must specify the lack of a unique limiting position. This error analysis question targets conceptual precision over binary true/false judgments.

📖 Explanation: Vector-valued continuity requires *all* component functions to be continuous simultaneously at the point. Even though the y-component is perfect and the z-component could be fixed by redefining a single point, the x-component’s jump discontinuity is irreparable and prevents the vector limit from matching the function value. This tests understanding that vector continuity is an 'all-or-nothing' property governed by the weakest link. Graph interpretation skills are needed to recognize that visual smoothness in some dimensions cannot compensate for failure in others. Students often mistakenly average behaviors or prioritize smoother components.

📖 Explanation: Indeterminate forms like $0/0$ in vector contexts often arise in kinematics when analyzing limiting positions during collisions or asymptotic approaches where both numerator and denominator vanish. Polynomials allow direct substitution. Steady-state problems typically involve PDEs, not single-variable vector limits. Arc length uses integration, not pointwise limits. Only option C presents a genuine modeling context where algebraic simplification or L'Hôpital's rule applied component-wise is necessary to resolve physical ambiguity. This connects abstract calculus techniques to tangible multi-step reasoning scenarios in dynamics.

📖 Explanation: This tests the chain of continuity: $\mathbf{r}$ continuous implies $\|\mathbf{r}\|$ continuous (norm is continuous everywhere including origin). Then $f$ continuous at that output implies the composition is continuous. Differentiability is irrelevant for mere continuity. The norm has no singularity affecting continuity; it is Lipschitz continuous globally. Many students falsely believe norms cause issues at zero, confusing continuity with differentiability. This conceptual question reinforces that continuity is preserved under composition regardless of geometric singularities that might affect derivatives. Direct recall of composition theorems suffices here.

📖 Explanation: Continuity holds because $|t^2 \sin(1/t)| \leq t^2 \to 0$. However, differentiability fails at 0 in the classical sense for navigation smoothness. Computing \mathbf{r}&#039;(t) for $t \neq 0$ gives terms involving $\cos(1/t)$ which oscillate without settling as $t \to 0$, so the derivative has no limit. Yet \mathbf{r}&#039;(0) exists via definition (equals 0). This creates a function that is differentiable but not continuously differentiable ($C^1$), explaining why motion feels jerky despite positional continuity. This mixed-concept problem links limits, continuity, and differentiability in applied modeling.

📖 Explanation: Standard limits give $(e^{kt}-1)/t \to k$ and $\tan(mt)/t \to m$, so first two components are correct. But $(1-\cos t)/t^2 \to 1/2$, not 0. Students often confuse this with $(1-\cos t)/t \to 0$ or forget the quadratic denominator scaling. This error analysis targets memory of fundamental trigonometric limits versus superficial pattern matching. Recognizing the Taylor expansion $\cos t \approx 1 - t^2/2$ immediately reveals the 1/2 factor. Multi-step verification of each component prevents cascading errors in vector limit evaluation.

📖 Explanation: This Olympiad-style question probes deep topological understanding. If the scalar projection converges for *every* direction $\mathbf{u}$, then in particular it converges for basis vectors, giving component limits. More profoundly, the condition actually characterizes weak convergence which in finite-dimensional spaces equals strong convergence. So yes, the vector limit exists. Distractors exploit confusion with multivariable scalar functions where directional limits don't suffice; but for vector-valued functions of a single variable, testing all directions is overkill yet sufficient. Basis vectors alone would suffice, making C tempting but unnecessarily restrictive. This tests abstraction beyond computational routines.

📖 Explanation: All components are continuous at 0 (absolute value cubed is still continuous). Derivatives: $d/dt(t^3)=3t^2 \to 0$; $d/dt(|t|^3) = 3t|t| \to 0$; $d/dt(t^2)=2t \to 0$. So \mathbf{r}&#039;(0) = \mathbf{0} exists. But second derivatives involve $|t|$ terms whose derivatives jump at 0, so \mathbf{r}&#039; is not continuous at 0. Hence differentiable but not $C^1$. This subtle distinction matters in motion planning where acceleration discontinuities cause mechanical stress. Students often conflate existence of first derivative with its continuity, especially with absolute values raised to odd powers.

📖 Explanation: Direct substitution gives $0/0$, but limits yield $\langle 1, 0, 0 \rangle$. Defining $\mathbf{v}(0)$ this way makes the velocity continuous, which is physically required for realistic fluid models (no infinite accelerations). Mass conservation relates to divergence, not pointwise continuity. Kinetic energy depends on speed squared, not definition choices. Turbulence involves Reynolds numbers, not single-point definitions. This application question connects removable discontinuities to physical plausibility, emphasizing that mathematical continuity corresponds to absence of unphysical impulses in dynamical systems.

📖 Explanation: L'Hôpital's rule for $\ln(1+2t)/t$ requires knowing $d/dx \ln(x) = 1/x$, whose proof often uses $\lim_{h \to 0} \ln(1+h)/h = 1$. Using L'Hôpital here can be circular depending on curriculum sequence. Conjugate and algebraic manipulation avoid this dependency. While both yield correct answers, Method B carries foundational risks in rigorous contexts. This comparison question develops metacognitive awareness of method selection beyond mere correctness, highlighting how procedural efficiency must be weighed against logical coherence in advanced mathematics education.

📖 Explanation: Although the curve oscillates densely between $y=x$ and $y=-x$, the amplitude $|t \sin(1/t)| \leq |t| \to 0$. Squeeze theorem guarantees convergence to origin regardless of visual complexity. Graphs can mislead by suggesting non-convergence when bounds tighten sufficiently. This contrasts with $\sin(1/t)$ alone which truly lacks a limit. Interpreting such plots requires analytical backup, not just visual intuition. This graph-based HOTS item trains students to distrust apparent chaos when dominating decay factors are present, reinforcing quantitative reasoning over perceptual judgment.

📖 Explanation: If $\mathbf{r}(t) \to \mathbf{0}$, then $\mathbf{r} \times \mathbf{s} \to \mathbf{0}$ holds even if $\mathbf{s}$ blows up, as long as $\mathbf{s}$ stays parallel to $\mathbf{r}$. Example: $\mathbf{r}=\langle t,0,0\rangle$, $\mathbf{s}=\langle 1/t,0,0\rangle$; cross product is always zero but $\mathbf{s}$ diverges. The student neglects that vanishing multiplier permits arbitrary growth in aligned directions. This error analysis exposes misunderstanding of indeterminate forms in vector products, analogous to scalar $0 \cdot \infty$. Recognizing degenerate cases is crucial for rigorous limit reasoning in higher dimensions.

📖 Explanation: The function $e^{-1/t^2}$ is classic smooth-but-not-analytic example. All components inherit this flatness at zero. Every derivative involves polynomials times $e^{-1/t^2}$, all tending to zero as $t \to 0$. Thus $\mathbf{r}$ is $C^\infty$ with identically zero Taylor series at origin despite being non-zero elsewhere. This challenges intuition linking smoothness to analyticity. Options test recognition of pathological smooth functions. Direct computation of first few derivatives suggests pattern; induction confirms infinite differentiability. This Olympiad-level item separates procedural fluency from deep structural understanding of function spaces.

📖 Explanation: Elementary functions (trig, polynomial, exponential) are continuous on their domains. Compositions and sums preserve continuity. Since cosine, sine, and identity are continuous on $\mathbb{R}$, their combinations are automatically continuous. Orbital closure relates to periodicity, not continuity. Constant parameters don't guarantee continuity if functions were discontinuous. Derivative existence implies continuity but isn't the reason here; continuity is more fundamental. This foundational recall anchors HOTS items by ensuring baseline knowledge before tackling complex scenarios. Recognizing automatic continuity saves unnecessary computation in applied settings.

📖 Explanation: Floating-point subtraction of nearly equal numbers (like $1-\cos(t)$ for tiny $t$) causes loss of significance. Series expansion $\cos(t) \approx 1 - t^2/2 + t^4/24$ avoids subtraction entirely. Reformulating as $2\sin^2(t/2)/t^2$ also helps. Brute-force precision increases cost without solving root cause. Symbolic tools help but aren't always available in embedded systems. This application question bridges pure math and computational reality, teaching that theoretical limits require numerically stable implementations. Understanding error sources is as important as knowing the analytical answer in scientific computing contexts.

📖 Explanation: Equal one-sided limits mean the two-sided limit exists and equals $\mathbf{L}$. Mismatch with function value defines removable discontinuity. Jump requires unequal one-sided limits. Infinite involves unboundedness. Essential covers oscillatory/non-existent limits. Vector-valued functions inherit scalar classification component-wise; if all components share this pattern, the vector has removable discontinuity. Redefining $\mathbf{r}(c) = \mathbf{L}$ restores continuity. This conceptual item ensures students classify discontinuities correctly in vector contexts, avoiding mislabeling based on single-component behavior or confusing existence of limit with equality to function value.

📖 Explanation: In $\mathbb{R}^n$, all norms are equivalent; convergence in Euclidean norm iff convergence in max-norm (component-wise). Specifically, $|r_i - L_i| \leq \|\mathbf{r}-\mathbf{L}\| \leq \sqrt{n} \max|r_i-L_i|$. So controlling norm controls components and vice versa. Components aren't always smaller (consider $\langle 3,4 \rangle$ vs components 3,4). Epsilon-delta works with any metric. Norm bounding implies component bounding but with potentially different epsilons due to constants. This mixed-concept item links topology, linear algebra, and analysis, showing why vector calculus can use unified norm arguments instead of tedious component tracking.

📖 Explanation: Norm is smooth away from origin but non-differentiable at zero. If $\mathbf{r}$ is continuous and $\|\mathbf{r}\|$ has a corner (non-differentiable point), likely $\mathbf{r}(t_0) = \mathbf{0}$. Elsewhere, composition of smooth functions remains smooth. Corners in distance functions signal passage through reference point. Discontinuity of $\mathbf{r}$ would cause worse singularities. Physics allows such geometric features. This synthesizes geometry, calculus, and modeling: interpreting non-smoothness in derived quantities to infer underlying state. Students must connect analytic properties to spatial configurations, moving beyond formula manipulation to physical insight.

📖 Explanation: Each option represents a documented misconception: arctan range confusion, ignoring exponential damping in oscillatory terms, and conflating vector limits with scalar magnitude limits. Correct evaluation gives $\langle \pi/2, 1, 0 \rangle$. Damped oscillation tends to zero by squeeze theorem. Rational function ratio of leading coefficients gives 1. Arctan asymptote is standard knowledge. Identifying multiple pitfalls strengthens diagnostic skills. This error-analysis item validates that mastery includes recognizing what *not* to do, preparing students for exam traps and real-world debugging where wrong assumptions propagate silently.

📖 Explanation: Near zero, smallest exponent dominates decay: $t^2 \gg t^3 \gg t^4$, so $\|\mathbf{r}\| \sim t^2$. At infinity, largest exponent dominates growth: $t^4 \gg t^3 \gg t^2$, so $\|\mathbf{r}\| \sim t^4$. Asymmetry arises because min/max roles reverse across regimes. Consecutive integers don't imply symmetry. This conceptual comparison builds intuition for asymptotic analysis in vector functions, essential for perturbation theory and scaling arguments. Students must identify dominant terms contextually, not memorize fixed hierarchies.

📖 Explanation: Implicit Function Theorem states that if $\mathbf{F}$ is $C^1$ and Jacobian w.r.t. $\mathbf{r}$ is nonsingular at solution point, then locally unique continuous (actually $C^1$) function $\mathbf{r}(t)$ exists. Mere continuity of $\mathbf{F}$ doesn't guarantee solvability. Differentiability in $t$ alone is insufficient. Boundedness doesn't imply continuity. This Olympiad-style item tests knowledge of existence theorems beyond explicit computation. Recognizing when implicit definitions yield well-behaved functions is crucial in differential equations and constrained optimization where closed-form solutions are unavailable.

📖 Explanation: Bernstein basis functions sum to 1 for all $t$, ensuring affine invariance and that curve lies in convex hull. At segment boundaries, shared control points and basis properties guarantee positional continuity ($C^0$) inherently. Endpoint interpolation is consequence, not cause. Piecewise polynomials can be discontinuous if segments don't match. Rendering enforcement is implementation detail, not mathematical guarantee. Partition of unity is the foundational property enabling robust geometric design. This application question connects abstract continuity concepts to industry-standard modeling, showing how mathematical structure replaces ad-hoc verification in engineered systems.