📝 Sum of infinite series in calculus (72 MCQs)

📖 Explanation: This question targets error analysis regarding the Divergence Test. A common misconception is assuming that $\lim_{k \to \infty} u_k = 0$ implies convergence. In reality, this limit being zero is merely a prerequisite; it does not guarantee a finite sum. The series in question is a p-series where $p = 1/2$. Since $p \leq 1$, the integral test confirms divergence despite the terms vanishing at infinity, illustrating that terms can shrink too slowly to produce a finite area.

📖 Explanation: This problem tests conceptual understanding of recursive limits. The validity of setting $\lim a_{n+1} = \lim a_n = L$ relies heavily on the continuity of the defining function. If the function were discontinuous at the limit point, this substitution would be unjustified. Furthermore, one must ensure the sequence actually converges (often via Monotone Convergence Theorem) before assuming $L$ exists. Simply solving the quadratic yields candidates, but continuity and convergence proofs validate that the specific positive root is indeed the true limit of the sequence.

📖 Explanation: This application question compares numerical methods. Integrating the Maclaurin series for $e^{-x^2}$ yields an alternating series. For such series, the error bound is simply the magnitude of the first omitted term, which is computationally cheap and exact. Conversely, Simpson’s Rule requires bounding the fourth derivative, which can be complex or large for this function. Additionally, series methods provide symbolic insight and arbitrary precision control, whereas numerical quadrature is limited by floating-point arithmetic and step-size constraints, making series preferable for rigorous high-accuracy demands in modeling.

📖 Explanation: This error analysis question addresses algebraic properties of divergent series. Students often incorrectly assume linearity extends to divergence. While the conclusion here is indeed correct (the partial sums oscillate indefinitely due to $(-1)^k$), the general principle cited is false. For example, $\sum \frac{1}{k}$ and $\sum -\frac{1}{k}$ both diverge, yet their sum converges to zero. Thus, one cannot deduce divergence of a sum solely from the divergence of components; one must analyze the combined behavior of partial sums directly.

📖 Explanation: This graph-based question interprets visual convergence behavior. The key feature is the approach to a horizontal asymptote, which defines convergence regardless of the decay rate, provided the limit exists. While $1/\ln(n)$ decays much slower than $1/n$, it still approaches zero, satisfying the definition of a limit. This highlights that convergence is a binary property (limit exists or not), distinct from the rate of convergence. Slow convergence has practical implications for computation but does not negate the theoretical existence of the sum, distinguishing mathematical truth from numerical feasibility.

📖 Explanation: This mixed-concept question links Taylor series to physical modeling. The first-order model assumes $\sin \theta \approx \theta$, implying period independence from amplitude (isochronism). The second-order term $k^2/4$ arises from the binomial expansion of the elliptic integral integrand. Its positive sign indicates that as initial angle $\theta_0$ (and thus $k$) increases, the period lengthens. This connects abstract series coefficients to tangible physical phenomena, showing how higher-order terms refine models by incorporating nonlinearities neglected in linearized approximations, crucial for accurate timekeeping and dynamics.

📖 Explanation: This conceptual question probes the limitations of convergence tests. The Ratio Test examines exponential growth rates. Since $(\frac{k}{k+1})^p \to 1$ for any finite $p$, the test sees all p-series as having the same 'boundary' growth rate. It lacks the sensitivity to detect polynomial differences in decay speed. This explains why the Integral Test, which directly measures area under $1/x^p$, is required: it distinguishes logarithmic vs. finite accumulation based on the exponent $p$, whereas ratio-based tests only resolve geometric-like behavior.

📖 Explanation: This application question emphasizes strategic problem-solving over brute force. Direct differentiation of $e^{-x^2}$ becomes rapidly intractable due to the chain rule and product rule complexity. However, since $e^u = \sum u^k/k!$ holds for all real $u$, substituting $u = -x^2$ immediately yields $\sum (-1)^k x^{2k}/k!$. Theorem 9.10.6 guarantees this resulting power series is indeed the Maclaurin series. This highlights the power of composition and substitution as primary tools for generating series, avoiding tedious calculus while maintaining rigor.

📖 Explanation: This challenging question explores the Riemann Rearrangement Theorem. The alternating harmonic series is the canonical example of conditional convergence. Because $\sum |a_k| = \sum 1/k$ diverges, the positive and negative parts each diverge to infinity independently. Convergence relies entirely on delicate cancellation in the original order. Rearranging disrupts this balance, allowing the partial sums to drift toward any target value or diverge. This contrasts sharply with absolutely convergent series, where commutativity holds. Understanding this distinction is vital for advanced analysis and prevents erroneous manipulation of conditionally convergent series in proofs.

📖 Explanation: This conceptual question clarifies the role of $M$ in error bounds. The theorem requires $|f^{(n+1)}(z)| \leq M$ for all $z$ between $x_0$ and $x$. Since sine and cosine derivatives are always bounded by 1 globally, choosing $M=1$ is always safe and avoids optimizing bounds for tiny intervals. While a tighter bound might reduce required terms slightly, $M=1$ guarantees correctness and is standard practice. This reinforces understanding that error bounds need only be valid upper limits, not necessarily sharp extrema, balancing rigor with computational efficiency.

📖 Explanation: This application question tests operational knowledge of power series manipulation. The factor $k$ in the numerator suggests differentiation, as $\frac{d}{dx}(x^k) = kx^{k-1}$. Starting from $\sum_{k=0}^\infty x^k = \frac{1}{1-x}$, differentiating both sides yields $\sum kx^{k-1} = \frac{1}{(1-x)^2}$. Crucially, this operation is valid only within the open interval of convergence $|x|<1$. This technique transforms known series into new ones efficiently, demonstrating how calculus operations extend the utility of basic series beyond direct summation formulas.

📖 Explanation: This error analysis question distinguishes between convergence testing and value computation. The Integral Test compares areas under curves to rectangle sums to determine finiteness, not equality. Geometrically, left-endpoint rectangles for a decreasing function overestimate the area, so $\sum a_k > \int f(x)dx$. Specifically, $\sum 1/k^2 = \pi^2/6 \neq 1$. Confusing the test's purpose (convergence criterion) with value calculation is a pervasive error. Clarifying this reinforces that integrals provide bounds and convergence status, while series sums require separate evaluation techniques like Fourier analysis or special identities.

📖 Explanation: This mixed-concept question connects abstract convergence domains to physical applicability. Bessel functions $J_0(x)$ and $J_1(x)$ arise in PDEs with cylindrical symmetry. An infinite radius of convergence means the power series representation is valid for all real $x$, corresponding to all possible radial distances in a physical system. If the radius were finite, the model would break down beyond some critical radius, rendering it useless for macroscopic problems. Thus, global convergence validates the series as a robust tool for describing physical fields throughout the entire domain without artificial boundaries.

📖 Explanation: This challenging question analyzes convergence acceleration. The standard alternating harmonic series decays as $O(1/n)$, requiring $n \approx 10^5$ for $10^{-5}$ error. The Gregory series substitutes $x=1/3$ into $\ln((1+x)/(1-x))$, introducing a factor $(1/3)^{2k+1}$. This geometric suppression $\sim 9^{-k}$ dominates the polynomial denominator, causing exponential error reduction. Each additional term adds roughly one digit of precision versus needing tenfold more terms for linear improvement. This exemplifies how variable transformations exploit analytic structure to convert impractical series into efficient computational algorithms.

📖 Explanation: This error analysis question targets misinterpretation of test outcomes. The Limit Comparison Test states that if $0 < \rho < \infty$, both series share the same fate. Since $\sum 1/k$ diverges, $\rho=1$ forces divergence of the given series. The student likely confused 'standard series' with 'convergent series' or misremembered the harmonic series' behavior. This highlights the necessity of knowing benchmark series properties cold. Correct application requires not just computing $\rho$ but correctly recalling whether the comparator converges or diverges, linking computational procedure to foundational knowledge.

📖 Explanation: This conceptual question addresses endpoint behavior under integration. The series $\sum (-1)^k x^{2k}$ for $1/(1+x^2)$ has radius 1 and diverges at $x=\pm 1$. However, integrating term-by-term yields $\sum (-1)^k x^{2k+1}/(2k+1)$, which satisfies the Alternating Series Test at $x=1$. Integration introduces a $1/n$ damping factor that can convert divergence to conditional convergence at boundary points. This illustrates that while radius of convergence remains unchanged under integration, endpoint behavior can improve, enabling evaluation of important constants like $\pi/4$ via series that originate from divergent parents.

📖 Explanation: This application question justifies practical truncation rules. For $e^x$ at $x=1$, the remainder $R_n = \sum_{k=n+1}^\infty 1/k!$. Factoring out $1/(n+1)!$ leaves a series bounded by a geometric series with ratio $1/(n+2) < 1$. Thus $R_n < \frac{1}{(n+1)!} \cdot \frac{n+2}{n+1} \approx \frac{1}{(n+1)!}$. Unlike alternating series where error < next term is exact, here it's an inequality derived from geometric domination. Understanding this bound validates using factorial decay as a proxy for total tail error, bridging theoretical remainders and computational stopping criteria.

📖 Explanation: This conceptual question probes the necessity of monotonicity in AST. Condition $\lim a_k = 0$ is necessary but insufficient; monotonic decrease ensures that successive partial sums nest within shrinking intervals. Without it, even if terms vanish, oscillations may persist with non-diminishing amplitude or erratic behavior preventing convergence. For example, interleaving $1/k$ and $1/k^2$ with alternating signs violates monotonicity and may diverge despite vanishing terms. This highlights that AST relies on structured cancellation enabled by ordered decay, not merely smallness of terms, distinguishing it from tests based solely on magnitude.

📖 Explanation: This mixed-concept question links series truncation to physical regimes. The binomial expansion parameter is $\beta^2 = v^2/c^2$. When $v \ll c$, $\beta^2 \approx 0$, making higher powers negligible. The first term $\frac{1}{2}m_0 v^2$ is precisely Newtonian kinetic energy. Subsequent terms are relativistic corrections scaling as $(v/c)^{2n}$. Thus, series expansion reveals Newtonian mechanics as the low-velocity asymptote of relativity. Truncation isn't mere approximation; it defines the domain of validity for classical physics. This exemplifies how Taylor series encode physical hierarchies, with each term representing progressively finer corrections to a base theory.

📖 Explanation: This graph-based question connects visual convergence rate to algorithmic efficiency. Gap size $|s_{n+1} - s_n| = |a_{n+1}|$ dictates error decay. Geometric decay halves error each step; achieving $10^{-6}$ needs ~20 terms. Harmonic decay reduces error inversely; same precision needs $10^6$ terms. Graphically, geometric envelopes collapse tightly; harmonic envelopes linger. Recognizing this visually informs method selection: slow-decaying series demand acceleration techniques or alternative representations. Thus, graphs aren't just verification tools; they diagnose computational feasibility, guiding practitioners toward tractable formulations before investing in brute-force summation.

📖 Explanation: This challenging question probes foundational analysis. In $\mathbb{Q}$, a bounded increasing sequence like decimal approximations of $\sqrt{2}$ has no rational limit; the supremum is missing. The Completeness Axiom asserts every nonempty bounded-above set in $\mathbb{R}$ has a supremum in $\mathbb{R}$. This fills gaps, ensuring monotone bounded sequences land on actual real numbers. Without it, convergence proofs fail because the candidate limit might not exist in the space. This axiom distinguishes $\mathbb{R}$ from $\mathbb{Q}$ and underpins all limit theorems in calculus, making it the bedrock of infinite series theory.

📖 Explanation: This application question highlights strategic efficiency. Computing $f^{(n)}(1)$ for $\ln x$ involves patterns like $(-1)^{n-1}(n-1)!$, derivable but tedious. Substituting $u=x-1$ into $\ln(1+u) = \sum (-1)^{k+1}u^k/k$ instantly gives $\sum (-1)^{k+1}(x-1)^k/k$. This exploits precomputed knowledge and algebraic simplicity. Moreover, it transfers known convergence properties ($-1<u\leq1 \Rightarrow 0<x\leq2$) without re-analysis. This exemplifies the broader principle: leverage existing series through transformation whenever possible. Direct computation is a fallback, not the default, emphasizing mathematical economy and pattern recognition over mechanical differentiation.

📖 Explanation: This conceptual question validates test selection. The Ratio Test excels with factorials and exponentials because $\frac{(k+1)!}{k!} = k+1$ and $\frac{(k+1)^{k+1}}{k^k}$ simplify cleanly. Here, $\frac{a_{k+1}}{a_k} = \frac{(k+1)!}{(k+1)^{k+1}} \cdot \frac{k^k}{k!} = \frac{k+1}{(k+1)^{k+1}} k^k = (1+\frac{1}{k})^{-k} \to 1/e$. Since $1/e < 1$, convergence is assured. The test handles mixed growth types effectively when ratios simplify to recognizable limits. This reinforces that Ratio Test applicability depends on algebraic tractability of the ratio, not term type exclusivity, making it versatile for combinatorial-exponential hybrids.

📖 Explanation: This error analysis question addresses unit consistency in calculus. Derivatives $\frac{d}{dx}\sin x = \cos x$ assume $x$ in radians because $\lim_{h\to0} \sin h / h = 1$ only in radians. In degrees, $\frac{d}{dx}\sin(x^\circ) = \frac{\pi}{180}\cos(x^\circ)$, altering all Taylor coefficients by powers of $\pi/180$. Using degrees in the standard series yields wildly incorrect values. This underscores that calculus-based series are inherently radian-dependent. Conversion isn't optional preprocessing; it's embedded in the series' derivation. Ignoring this breaks the mathematical model, highlighting the importance of dimensional awareness in applied analysis.

📖 Explanation: This conceptual question tests deep understanding of convergence types. Absolute convergence ($\sum |a_k| < \infty$) implies unconditional convergence; commutativity holds as in finite sums. Conditional convergence ($\sum a_k$ converges but $\sum |a_k|$ diverges) relies on cancellation order. Riemann's Rearrangement Theorem shows such series are maximally unstable: terms can be reordered to yield any prescribed sum or diverge. This dichotomy reveals that absolute convergence reflects intrinsic summability, while conditional convergence is an artifact of sequencing. Recognizing this prevents dangerous manipulations and clarifies why absolute convergence is the gold standard in analysis and applications.

📖 Explanation: This challenging question dissects proof mechanics. The double inequality bounds $s_n$ above and below by integrals differing only by finite terms/constants. If the integral converges, the upper bound is finite, so $s_n$ is bounded above and increasing ⇒ converges. If the integral diverges, the lower bound → ∞ ⇒ $s_n$ → ∞. A single-sided comparison couldn't establish both directions of equivalence. The sandwich structure is what makes the test bidirectional (converge⇔converge, diverge⇔diverge). Understanding this reveals why precise geometric alignment of rectangles and curves matters: it creates the logical bridge between discrete sums and continuous integrals.

📖 Explanation: This mixed-concept question applies series to dynamical systems. Near equilibrium $L$, write $x_n = L + \epsilon_n$. Taylor expanding f(L+\epsilon) \approx L + f&#039;(L)\epsilon gives \epsilon_{n+1} \approx f&#039;(L)\epsilon_n. This linear approximation governs local dynamics: |f&#039;(L)|<1 ⇒ exponential decay to $L$; $>1$ ⇒ instability. Higher-order terms matter only far from equilibrium. Thus, Taylor series reduce complex nonlinear stability analysis to eigenvalue examination. This exemplifies how series serve as microscopes for local behavior, transforming intractable recurrences into tractable linear models near critical points, foundational in ecology and physics.

📖 Explanation: This conceptual question explains binomial series structure. The general coefficient is $\binom{m}{k} = \frac{m(m-1)\cdots(m-k+1)}{k!}$. If $m \in \mathbb{N}_0$, when $k = m+1$, the numerator includes $(m-(m+1)+1) = 0$, annihilating all higher terms. This recovers the finite Binomial Theorem. For non-integer $m$, no factor ever vanishes, producing infinite series converging for $|x|<1$. This unifies finite algebra and infinite analysis under one formula. Recognizing termination conditions prevents misapplying infinite series machinery to polynomials and clarifies the boundary between algebraic identities and analytic expansions.

📖 Explanation: This application question highlights series advantages for singular integrands. At $x=0$, $\sin x/x$ is undefined but has limit 1. Numerical methods may fail or require special handling at this point. The Maclaurin series $\sum (-1)^k x^{2k}/(2k+1)!$ is defined and smooth at 0, integrating to $\sum (-1)^k /[(2k+1)(2k+1)!]$. This alternating series converges rapidly with simple error bounds. Series thus regularizes the singularity naturally, providing stable, high-precision results where naive quadrature struggles. This demonstrates series as analytical preprocessors that transform problematic integrals into well-behaved sums, combining robustness with accuracy.

📖 Explanation: This challenging question tests nuanced convergence relationships. Absolute convergence ($\sum |a_k| < \infty$) implies $|a_k| \to 0$, so for large $k$, $|a_k| < 1$ ⇒ $a_k^2 < |a_k|$. Comparison Test then gives $\sum a_k^2$ convergence. But conditional convergence doesn't guarantee this: e.g., $a_k = (-1)^k/\sqrt{k}$ converges conditionally, but $a_k^2 = 1/k$ diverges. Thus, the inference requires absolute convergence or positivity. This reveals that squaring amplifies small terms differently based on sign structure, making absolute convergence the sufficient safeguard. Understanding this prevents overgeneralization from specific examples.

📖 Explanation: This application question emphasizes limit evaluation skills. Applying Root Test: $\sqrt[k]{|(1+1/k)^{k^2} x^k|} = (1+1/k)^k |x| \to e|x|$. Convergence requires $e|x| < 1$ ⇒ $|x| < 1/e$. Mis-evaluating $(1+1/k)^k \to 1$ would wrongly suggest $R=1$. This limit is a cornerstone of analysis; its appearance in series tests is frequent. Recognizing it transforms ambiguous expressions into precise radii. This underscores that convergence tests often hinge on standard limits; mastery of these is as vital as knowing the tests themselves, linking sequential and exponential growth concepts.

📖 Explanation: This Olympiad-style question explores pathological functions. Despite $C^\infty$ smoothness, $f^{(n)}(0) = 0$ for all $n$, so Taylor series is $\sum 0 \cdot x^n = 0$. This converges everywhere but matches $f(x)$ only at origin. Such functions are smooth but non-analytic, revealing that Taylor series represent only analytic functions. Infinite differentiability is necessary but not sufficient for Taylor equality; remainder $R_n(x)$ must → 0. This counterexample shatters the intuition that smoothness implies series representability, highlighting the subtle distinction between $C^\infty$ and analytic classes central to advanced analysis.

📖 Explanation: This error analysis question addresses discrete vs. continuous tools. $n!$ lacks a derivative, blocking direct L'Hôpital. However, writing $n!/n^n = (1/n)(2/n)\cdots(n/n) \leq (1/n) \cdot 1 \cdots 1 = 1/n$ creates bounds $0 \leq a_n \leq 1/n$. Both bounds → 0, forcing $a_n \to 0$. This leverages discrete structure inaccessible to calculus. Squeezing excels where continuous methods fail, especially with factorials, products, or oscillatory terms. Recognizing when to switch from calculus to inequality-based reasoning is crucial for sequence limits, showcasing the complementary roles of analytic and algebraic techniques in convergence proofs.

📖 Explanation: This mixed-concept question addresses approximation validity domains. Small f&#039;&#039;&#039;(x_0) ensures good local fit near $x_0$, but accuracy degrades with distance as $|x-x_0|^3$ grows. Over extended intervals, even tiny cubic coefficients accumulate significant error. Also, higher derivatives may grow rapidly away from $x_0$, invalidating local bounds. Thus, sufficiency depends on both derivative magnitude and interval width relative to convergence radius. This highlights that Taylor approximations are inherently local; extrapolation risks require global analysis or adaptive methods. Modeling demands matching approximation order to operational domain, not just pointwise derivative size.

📖 Explanation: This conceptual question reinforces procedural completeness. Ratio Test yields $\rho = |x-a|/R$; convergence for $\rho<1$, divergence for $\rho>1$, but $\rho=1$ is indeterminate. Endpoint behavior varies: $\sum x^k/k^2$ converges at both; $\sum x^k/k$ converges at -1, diverges at 1; $\sum x^k$ diverges at both. Assuming uniformity risks incorrect domains. Separate tests (AST, p-series, etc.) are mandatory at boundaries. This step ensures precise convergence sets, critical for function representation validity. Neglecting it is a common error that undermines the rigor of power series analysis.

📖 Explanation: This question targets error analysis regarding the distinction between the limit of the general term $u_k$ and the sum of the series $S$. A common misconception is believing $\lim_{k \to \infty} u_k = 0$ implies $\sum u_k = 0$ or even just convergence. In reality, $\lim u_k = 0$ is a necessary condition for convergence but says nothing about the value of the sum. The sum is defined as the limit of the sequence of partial sums $s_n$. For this specific telescoping series, the partial sums simplify to $1 - \frac{1}{n+1}$, which converges to 1, not 0. Understanding this distinction is critical for mastering infinite series.

📖 Explanation: This application question connects the abstract concept of infinite geometric series to the concrete representation of rational numbers. The repeating decimal $0.1212...$ can be written as the series $\frac{12}{100} + \frac{12}{10000} + \dots$. Here, the first term is $a = \frac{12}{100}$ and the common ratio is $r = \frac{1}{100}$. Using the sum formula $S = \frac{a}{1-r}$, we get $\frac{12/100}{1 - 1/100} = \frac{12/100}{99/100} = \frac{12}{99}$. This simplifies to $\frac{4}{33}$. Option B represents a truncation error, while C and D arise from incorrect identification of the ratio or first term. This reinforces modeling real numbers via series.

📖 Explanation: This challenging problem requires analyzing a series where terms vanish but the sum explodes. While the Divergence Test is inconclusive (limit is 0), recognizing the telescoping structure is key. Writing out partial sums: $s_1 = \sqrt{2}-\sqrt{1}$, $s_2 = \sqrt{3}-\sqrt{1}$, ..., $s_n = \sqrt{n+1}-1$. Taking the limit as $n \to \infty$, $\sqrt{n+1} \to \infty$, so $s_n \to \infty$. This highlights that vanishing terms do not guarantee a finite sum. Distractors involve misapplying tests: Integral Test actually confirms divergence (integral of derivative-like term grows), and Ratio Test is indeed inconclusive. Only the direct evaluation of partial sums provides definitive proof here.

📖 Explanation: This scenario-based question tests conceptual understanding of modeling physical phenomena with series. Students often blindly apply $\frac{a}{1-r}$ without considering the physical context. The total distance includes the initial fall (10m) plus the up-and-down travel for each bounce. The height of the first bounce is $10(3/4)$, contributing $2 \times 10(3/4)$ to distance. Thus, total distance = $10 + 2[10(3/4) + 10(3/4)^2 + ...]$. Option B confuses distance with time. Option C introduces irrelevant physics. Option D is nonsensical. Correctly setting up the series based on the physical description is a crucial HOTS skill distinguishing rote memorization from true comprehension.

📖 Explanation: This graph-based question requires interpreting visual data about partial sums. Oscillating partial sums with decreasing amplitude converging to a limit $L$ strongly suggest an alternating series or similar behavior. Convergence is defined by $\lim s_n = L$, so option B is false. Absolute convergence (C) cannot be deduced solely from conditional convergence visuals; $\sum (-1)^k/k$ fits the graph but isn't absolutely convergent. While geometric series (D) can look like this, many non-geometric alternating series do too. Thus, A is the only definitive conclusion. This tests understanding that convergence depends on the limit of partial sums, not their exact path, and distinguishes conditional from absolute convergence visually.

📖 Explanation: This error analysis question addresses a subtle pitfall in telescoping series. When the gap between subtracted terms is greater than 1 (here, $k$ vs $k+2$), multiple initial AND trailing terms persist. Expanding: $(1 - 1/3) + (1/2 - 1/4) + (1/3 - 1/5) + ...$. Terms $1/3, 1/4, ...$ cancel, but $1$ and $1/2$ remain at the start, and $-1/(n+1)$ and $-1/(n+2)$ remain at the end. As $n \to \infty$, trailing terms go to 0, leaving $1 + 1/2 = 3/2$. Students often assume only one term survives. This multi-step reasoning ensures deep understanding of partial sum mechanics beyond simple pattern matching.

📖 Explanation: This conceptual question probes the counterintuitive nature of the harmonic series. Many students struggle because $1/k \to 0$. The classic Oresme proof groups terms: $1 + 1/2 + (1/3+1/4) + (1/5+...+1/8) + ... > 1 + 1/2 + 1/2 + 1/2 + ...$, showing unbounded growth. Option B is factually wrong ($1/k \neq 1/\ln k$). Option C misstates p-series theory (decreasing p makes it diverge more). Option D invokes computational artifacts irrelevant to pure math. Understanding that 'terms going to zero' is insufficient and that cumulative growth rate matters is fundamental. This distinguishes procedural knowledge from conceptual grasp of convergence criteria.

📖 Explanation: This mixed-concept problem combines algebraic manipulation with geometric series properties. We have two equations: $\frac{a}{1-r} = 6$ and $\frac{a}{1-r^2} = 4$. Dividing the first by the second gives $\frac{1-r^2}{1-r} = \frac{6}{4} \Rightarrow 1+r = \frac{3}{2} \Rightarrow r = \frac{1}{2}$. Wait, checking options: if $r=1/2$, then $a/(1-1/2)=6 \Rightarrow 2a=6 \Rightarrow a=3$. But let's re-evaluate the system. Actually, $\frac{a}{1-r}=6$ and $\frac{a}{(1-r)(1+r)}=4$. Substitute $a=6(1-r)$ into second: $\frac{6(1-r)}{(1-r)(1+r)}=4 \Rightarrow \frac{6}{1+r}=4 \Rightarrow 1+r=1.5 \Rightarrow r=0.5$. Then $a=3$. Option B matches. Let me re-check Option A. If $a=9/2, r=1/3$: Sum1 = $(9/2)/(2/3) = 27/4 = 6.75 \neq 6$. So B is correct. This requires solving a nonlinear system derived from series formulas, testing algebraic fluency alongside series knowledge.

📖 Explanation: This application question tests practical use of the Alternating Series Estimation Theorem. For a convergent alternating series satisfying standard conditions, $|S - s_n| \leq a_{n+1}$. Here $a_{n+1} = \frac{1}{n+1}$. Setting $\frac{1}{n+1} < 0.001$ gives $n+1 > 1000 \Rightarrow n \geq 999$. So 1000 terms suffice. Option B wrongly assumes quadratic convergence. Option C misunderstands error bounds. Option D falsely claims geometric convergence. This highlights the slow convergence of the alternating harmonic series and reinforces that error estimation relies on the next term, not asymptotic rates. It bridges theoretical bounds with computational reality, a key HOTS application.

📖 Explanation: This error analysis question targets misuse of linearity properties. Linearity $\sum (a_k + b_k) = \sum a_k + \sum b_k$ holds ONLY if both individual series converge. Here, $\frac{2k+3}{k(k+1)} = \frac{3}{k} - \frac{1}{k+1}$ via partial fractions. Both $\sum 3/k$ and $\sum -1/(k+1)$ diverge, so you cannot split them. Even the student's incorrect split has a divergent part. Recognizing when operations are invalid is as important as knowing when they work. This prevents blind application of rules. The correct approach treats it as a single telescoping entity or recognizes dominant harmonic behavior. This tests meta-cognitive awareness of theorem prerequisites.

📖 Explanation: This mixed-concept question integrates alternating series, absolute convergence, and p-series. First, check alternating series conditions: $1/\sqrt{k}$ is positive, decreasing, and limits to 0, so it converges. Next, check absolute convergence: $\sum |(-1)^k/\sqrt{k}| = \sum 1/k^{0.5}$. This is a p-series with $p=0.5 \leq 1$, so it diverges. Since the original converges but absolute version diverges, it's conditionally convergent. Option B is false (absolute divergence). Option C is false (monotonicity holds). Option D contradicts itself. This requires synthesizing multiple tests and definitions, moving beyond single-test application to holistic classification, a hallmark of higher-order thinking.

📖 Explanation: This challenging modeling question links recursive sequences to series concepts. Near equilibrium $L$, linearizing the Beverton-Holt map shows $x_{n+1} - L \approx c(x_n - L)$ where $|c|<1$. Thus deviations decay geometrically. The total 'adjustment' or cumulative difference relates to summing these geometric deviations. Option B suggests harmonic decay (too slow for stable fixed points). Option C implies finite steps or specific cancellation not generally true. Option D suggests oscillation, but Beverton-Holt is monotonic for positive parameters. Recognizing that local stability implies geometric contraction connects dynamical systems to series convergence. This transcends textbook problems, applying series theory to biological models, demanding deep conceptual transfer.

📖 Explanation: This conceptual question addresses the profound difference between absolute and conditional convergence. Absolute convergence ($\sum |a_k| < \infty$) guarantees rearrangement invariance. Conditional convergence does not. $\sum (-1)^{k+1}/k$ converges conditionally (harmonic abs diverges), so Riemann's theorem applies. $\sum (-1)^{k+1}/k^2$ converges absolutely (p=2>1), so rearrangement is safe. Option B falsely claims $1/k^2$ is conditional. Option C reverses the classifications. Option D is universally false. Understanding that 'convergence' alone doesn't imply 'robustness' is critical. This challenges the intuitive notion that addition is always commutative, extending finite arithmetic intuition carefully into the infinite realm, a key HOTS conceptual hurdle.

📖 Explanation: This application question reverses the usual process: given $s_n$, find convergence and terms. By definition, if $\lim s_n = S$ exists, the series converges to $S$. Here $\lim \frac{3n^2+2n}{4n^2+1} = 3/4$, so sum is $3/4$. Also $u_k = s_k - s_{k-1}$. Note $\lim u_k = \lim (s_k - s_{k-1}) = 3/4 - 3/4 = 0$, satisfying necessary condition. Option B is wrong (terms DO go to 0). Option C confuses term limit with sum. Option D uses calculus incorrectly (discrete vs continuous). This tests foundational definitions: sum IS limit of partial sums. Reinforces that $s_n$ encapsulates all series information, promoting flexible thinking between sequences and series.

📖 Explanation: This error analysis question tackles asymptotic reasoning misconceptions. While $\sin(x) \sim x$ is valid, it means $\sin(1/k)$ and $1/k$ share convergence/divergence behavior via Limit Comparison Test. Since $\sum 1/k$ diverges, so does $\sum \sin(1/k)$. 'Almost converges' reflects fuzzy thinking; divergence is binary. Option B misapplies comparison (smaller than divergent tells nothing). Option C is factually wrong ($\sin(1/k)>0$ for $k \geq 1$). Option D is irrelevant (calculus assumes radians). This emphasizes rigorous application of limit comparison over intuitive 'closeness'. Students must understand that asymptotic equivalence preserves divergence just as it preserves convergence, correcting a pervasive cognitive bias.

📖 Explanation: This graph interpretation question links visual trends to term properties. If $s_n$ approaches $L$ from ABOVE with DECREASING step sizes, then $s_{n} - s_{n-1} = u_n < 0$ (since decreasing) and $|u_n| = s_{n-1} - s_n$ is getting smaller (steps shrinking). So $u_k$ are negative with decreasing absolute value. Option B describes oscillation (not monotonic approach). Option C contradicts the graph direction. Option D says magnitudes increase, which would mean steps get larger, contradicting 'each step down smaller'. Interpreting discrete graphs requires translating visual slope/position into algebraic term properties. This spatial-to-symbolic reasoning is essential for deep series comprehension beyond formula plugging.